ANOVAs_in_R.ipynb

{
 "worksheets": [
  {
   "cells": [
    {
     "metadata": {},
     "cell_type": "heading",
     "source": "ANOVAs in R",
     "level": 2
    },
    {
     "metadata": {},
     "cell_type": "markdown",
     "source": "I've spent much longer than I would have liked to looking up how do a mixed within + between factors ANOVA in R. \n\nMaking some notes of what I've found here for prosperity. "
    },
    {
     "metadata": {},
     "cell_type": "markdown",
     "source": "There are several packages that implement various kind of ANOVA functionality:\n    \n - aov\n - car\n - ezANOVA\n - lmer (lme4)\n - lme (nlme)\n - BayesFactor\n    "
    },
    {
     "metadata": {},
     "cell_type": "markdown",
     "source": "I have concentrated on 'aov' , 'lmer' and, somewhat tangentially, BayesFactor.\n\nThe following is a collection of specific and general points, and then some useful code examples. "
    },
    {
     "metadata": {},
     "cell_type": "heading",
     "source": "Random tidbids",
     "level": 2
    },
    {
     "metadata": {},
     "cell_type": "markdown",
     "source": "Here are a list of interesting / useful / relevant / fuck knows comments from the interwebIn my scouting I have found. \n\nIt seemed worth jotting them down."
    },
    {
     "metadata": {},
     "cell_type": "markdown",
     "source": "[Excellent website for statistical R code examples](http://www.uni-kiel.de/psychologie/rexrepos/posts/anovaMixed.html)\n\n\n...similarly: ['R cookbook'](http://www.cookbook-r.com/Statistical_analysis/ANOVA/)\n\nR codeshool website from Rog\nhttp://tryr.codeschool.com/levels/1/challenges/1\n\n\n\nPhd comics on ANOVA\nhttp://www.phdcomics.com/comics/archive.php?comicid=905\n\n"
    },
    {
     "metadata": {},
     "cell_type": "heading",
     "source": "*lme4 doesn't give p values*",
     "level": 3
    },
    {
     "metadata": {},
     "cell_type": "markdown",
     "source": "This apparently is important, and according to Douglas Bates and others, in many cases when p values are provided they are wrong. "
    },
    {
     "metadata": {},
     "cell_type": "markdown",
     "source": "[rant on R list](https://stat.ethz.ch/pipermail/r-help/2006-May/094765.html)\n\n[discussion in post on mixed models](https://stat.ethz.ch/pipermail/r-sig-mixed-models/2008q2/000904.html)\n    \n[another post discussion](http://stats.stackexchange.com/questions/22988/significant-effect-in-lme4-mixed-model)\n        \n"
    },
    {
     "metadata": {},
     "cell_type": "markdown",
     "source": "Some posts on stack exchange suggest using 'pval.func' from the 'LanguageR' package to compute p values for lme4 models. \n\nBut apparently this is problematic and the lme4 authors don't appear to be inclined to fix it; \n\n[lm4-languageR incompatibility discussion](http://stackoverflow.com/questions/19199713/lme4-and-languager-compatibility-error-input-model-is-not-a-mer-object)\n\n[similar query along those lines](http://stackoverflow.com/questions/19568597/how-can-i-run-lme4-with-pvals-fnc-to-compute-p-value-in-mavericks)\n\n"
    },
    {
     "metadata": {},
     "cell_type": "markdown",
     "source": "...see also the 'pvalues' section of the [lme4 manual](http://cran.r-project.org/web/packages/lme4/lme4.pdf):\n\n\n>One of the most frequently asked questions about\n  lme4 is \"how do I calculate p-values for estimated\n  parameters?\" Previous versions of lme4 provided\n  the mcmcsamp function, which efficiently generated\n  a Markov chain Monte Carlo sample from the posterior\n  distribution of the parameters, assuming flat (scaled\n  likelihood) priors. Due to difficulty in constructing a\n  version of mcmcsamp that was reliable even in\n  cases where the estimated random effect variances were\n  near zero (e.g. https://stat.ethz.ch/pipermail/r-sig-mixed-models/2009q4/003115.html}),\n  mcmcsamp has been withdrawn (or more precisely,\n  not updated to work with \\code{lme4} versions >=1.0.0).\n  Many users, including users of the aovlmer.fnc function from the languageR package which relies\n  on mcmcsamp, will be deeply disappointed by this lacuna. Users who need p-values have a variety\n  of options:  \n  \n>  • likelihood ratio tests via anova (MC,+)  \n    • profile confidence intervals via profile.merMod and confint.merMod (CI,+)  \n    • parametric bootstrap confidence intervals and model comparisons via bootMer (or PBmodcomp in the pbkrtest package) (MC/CI,*,+)  \n    • for random effects, simulation tests via the RLRsim package (MC,*)  \n    • for fixed effects, F tests via Kenward-Roger approximation using KRmodcomp from the pbkrtest package (MC)  \n    • car::Anova and lmerTest::anova provide wrappers for pbkrtest: the latter also provides t tests via the Satterthwaite   approximation (P,*)  \n    \n>In the list above, the methods marked MC provide explicit model comparisons; CI denotes confidence\nintervals; and P denotes parameter-level or sequential tests of all effects in a model. The starred (*)\nsuggestions provide finite-size corrections (important when the number of groups is <50); those\nmarked (+) support GLMMs as well as LMMs.\nWhen all else fails, don’t forget to keep p-values in perspective: http://www.phdcomics.com/comics/archive.php?comicid=905\n\n  "
    },
    {
     "metadata": {},
     "cell_type": "markdown",
     "source": "One option mentioned above and [here](http://stackoverflow.com/questions/19199713/lme4-and-languager-compatibility-error-input-model-is-not-a-mer-object) is the is the Satterthwaite approximation in the ['lmerTest' package](http://cran.r-project.org/web/packages/lmerTest/lmerTest.pdf). That's what I'm currently looking into. "
    },
    {
     "metadata": {},
     "cell_type": "heading",
     "source": "*Other random bits and bobs*",
     "level": 3
    },
    {
     "metadata": {},
     "cell_type": "markdown",
     "source": "[Someone saying that lmer doesn't work for repeated measures ANVOA](http://permalink.gmane.org/gmane.comp.lang.r.lme4.devel/9991)\n(???)\n\n\n([Why does a mixed design using R's aov() need the between subject factors specified more than once?](http://stats.stackexchange.com/questions/45264/why-does-a-mixed-design-using-rs-aov-need-the-between-subject-factors-specifi)  \n  (...points to an error in the cookbook...)\n\n    \n[aov / lmer repeated measures ANOVA code conversion question](http://stackoverflow.com/questions/15131977/repeated-measure-anova-using-regression-models-lm-lmer?rq=1)    \n\n[repeated meaures within subejcts anova in R question](http://stackoverflow.com/questions/5694664/repeated-measures-within-subjects-anova-in-r)  \n...reply states that lme4 is more popular than nlme or aov these days...\n\n[Difference between 2 factor anova and mixed effects model](http://stats.stackexchange.com/questions/56380/difference-between-a-2-factor-anova-and-mixed-effects-model?rq=1)\n\n[more qs about lme4 p values](http://stats.stackexchange.com/questions/22988/significant-effect-in-lme4-mixed-model)\n\n\n[post hoc test for between subject factor in repeated meaures anova](http://stats.stackexchange.com/questions/52583/post-hoc-test-for-between-subject-factor-in-a-repeated-measures-anova-in-r?rq=1)\n\n[Repeated measures ANOVA with lme for 2 within subjects factors](http://stats.stackexchange.com/questions/13784/repeated-measures-anova-with-lme-in-r-for-two-within-subject-factors)\n\n\n[specifying lme models with more than one wihtin subjects factor](http://stats.stackexchange.com/questions/23833/is-there-a-way-to-specify-a-lme-model-with-more-than-one-within-subjects-factor?rq=1)\n\n[repeated measures anova in R](http://www.r-statistics.com/2010/04/repeated-measures-anova-with-r-tutorials/)  \n(blog post; but doesn't talk about the packages I'm looking at)\n    \n\n[more on lmr and aov](https://stat.ethz.ch/pipermail/r-help/2006-August/110788.html)  \n\n[anatomy of a mixed model in lme4 (PDF slides)](http://maths-people.anu.edu.au/~johnm/r-book/xtras/mlm-ohp.pdf)     \n\n\n[lme4 tutorial documentation](http://lme4.r-forge.r-project.org/lMMwR/lrgprt.pdf)\n    \n    \n[Multivariate linear modelling with car package](http://journal.r-project.org/archive/2013-1/fox-friendly-weisberg.pdf)\n    \n    \n[Bayesfactor package](http://bayesfactorpcl.r-forge.r-project.org/#mixed)\n(reasonably sophisticated section on ANOVA)\n    \n    "
    },
    {
     "metadata": {},
     "cell_type": "markdown",
     "source": "from here http://stats.stackexchange.com/questions/45264/why-does-a-mixed-design-using-rs-aov-need-the-between-subject-factors-specifi  \n...(after noting several incorrectly specified examples on line )"
    },
    {
     "metadata": {},
     "cell_type": "heading",
     "source": "Example code",
     "level": 2
    },
    {
     "metadata": {},
     "cell_type": "heading",
     "source": "*Various function calls from different packages*",
     "level": 4
    },
    {
     "metadata": {},
     "cell_type": "markdown",
     "source": "aov calls:\n\n```R\nfm <- aov(yield ~ v + n*p*k + Error(farms/blocks), data=farm.data)\n```\n\n\naov with 3 within subjects factors\n\n```R\n%R m.aov<-aov(measure~(task*region*actiontype) + Error(subject/(task*region*actiontype)),data)\n```\n\n\n2 factor ANOVA\n```R\nsummary(aov(angle ~ temperature + recipe, cake))\n```"
    },
    {
     "metadata": {},
     "cell_type": "markdown",
     "source": "lmer from lme4 package calls:\n\n\nmixed effects model with temperature as a random effect\n```R\nlmer(angle ~ recipe + (1| temperature), data=cake, REML=F)\n```\n\n\n(?)\n\n```R\nlmer(Y ~ A*B + (1|subject) + (1|A:subject) + (1|B:subject), data=d)  \n```"
    },
    {
     "metadata": {},
     "cell_type": "markdown",
     "source": "lme from nlme:\n\n\n```R\nlme(Y ~ A*B, random=list(subject=pdBlocked(list(~1, pdIdent(~A-1), pdIdent(~B-1)))), data=d)\n```"
    },
    {
     "metadata": {},
     "cell_type": "markdown",
     "source": "ez anova package:\n\nwith between subjects X and within subejcts X\n```R\nezANOVA(data=NoP_3_5,dv=.(AcPO4),wid=.(Subject),within=.(Time),between=.(Species),return_aov=T)\n$ANOVA\n```"
    },
    {
     "metadata": {},
     "cell_type": "heading",
     "source": "My example analyses",
     "level": 2
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "%load_ext rmagic\n\n%R library(lme4Test)\n\n%R library(BayesFactor)",
     "prompt_number": 6,
     "outputs": [
      {
       "output_type": "stream",
       "text": "Error in library(lme4Test) : there is no package called ‘lme4Test’\n",
       "stream": "stdout"
      },
      {
       "text": "Loading required package: coda\nLoading required package: lattice\n************\nWelcome to BayesFactor 0.9.6. If you have questions, please contact Richard Morey ([email protected]).\n\nType BFManual() to open the manual.\n************\n",
       "output_type": "display_data",
       "metadata": {}
      }
     ],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "import pandas as pd",
     "prompt_number": 1,
     "outputs": [],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "df_oc = pd.read_pickle('../DoctoralThesis/Chapter5/data/df_ols_categorized.pkl')",
     "prompt_number": 2,
     "outputs": [],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "#import pandas as pd\n#df_oc =pd.read_pickle('data/df_ols_categorized.pkl')\n\nhemilabs = []\nfor i in df_oc.index:\n if 'L' in df_oc.ix[i].ROI:\n  hemilabs.append('L')\n elif 'R' in df_oc.ix[i].ROI:\n  hemilabs.append('R')\ndf_oc['hemi'] = hemilabs        \n\ndf_oc_vis = df_oc[df_oc.modality=='vid']\ndf_oc_aud = df_oc[df_oc.modality=='aud']\ndf_oc_LH = df_oc[df_oc.hemi=='L']\ndf_oc_RH = df_oc[df_oc.hemi=='R']\n\ndf_oc_vis_LH = df_oc_vis[df_oc_vis.hemi=='L']\ndf_oc_vis_RH = df_oc_vis[df_oc_vis.hemi=='R']\ndf_oc_aud_LH = df_oc_aud[df_oc_aud.hemi=='L']\ndf_oc_aud_RH = df_oc_aud[df_oc_aud.hemi=='R']\n\n# can't use merge because the identifying things aren't indixes\n#df_oc_vis_LH = pd.merge(df_oc_vis, df_oc_LH, how='outer')",
     "prompt_number": 3,
     "outputs": [],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "aov_1a_str = \"\"\"dat ~ AgeGroup * withinORcross * modality * hemi + Error(CCID/(AgeGroup * withinORcross * modality * hemi))\"\"\"\naov_1b_str = 'dat ~ AgeGroup * withinORcross * modality  + Error(CCID/(AgeGroup * withinORcross * modality ))'\naov_1c_str = 'dat ~ AgeGroup * withinORcross * hemi + Error(CCID/(AgeGroup * withinORcross * hemi))'\naov_1d_str = 'dat ~ AgeGroup * withinORcross * + Error(CCID/(AgeGroup * withinORcross))'\naov_1e_str = 'dat ~ AgeGroup + Error(CCID/(AgeGroup ))'",
     "prompt_number": 4,
     "outputs": [],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "aov_1a_str = \\\n'aov_1a <- aov(dat ~ AgeGroup * withinORcross * modality * hemi + Error(CCID/(AgeGroup * withinORcross * modality * hemi)),data=df_oc)'\n\naov_1b_LH_str = \\\n'aov_1b_LH <- aov(dat ~ AgeGroup * withinORcross * modality  + Error(CCID/(AgeGroup * withinORcross * modality )),data=df_oc_LH)'\n\naov_1b_RH_str = \\\n'aov_1b_RH <- aov(dat ~ AgeGroup * withinORcross * modality  + Error(CCID/(AgeGroup * withinORcross * modality )),data=df_oc_RH)'\n\n\n\n#aov_1b_RH_str = 'dat ~ AgeGroup * withinORcross * modality  + Error(CCID/(AgeGroup * withinORcross * modality ))'\n#aov_1c_str = 'dat ~ AgeGroup * withinORcross * hemi + Error(CCID/(AgeGroup * withinORcross * hemi))'\n#aov_1d_str = 'dat ~ AgeGroup * withinORcross * + Error(CCID/(AgeGroup * withinORcross))'\n#aov_1e_str = 'dat ~ AgeGroup + Error(CCID/(AgeGroup ))'",
     "prompt_number": 5,
     "outputs": [],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {},
     "cell_type": "markdown",
     "source": "http://www.uni-kiel.de/psychologie/rexrepos/posts/anovaMixed.html#two-way-repeated-measures-anova-rbf-pq-design"
    },
    {
     "metadata": {},
     "cell_type": "markdown",
     "source": "Three-way split-plot factorial ANOVAs\n\n\naov:\n\n```R\nsummary(aov(Y ~ Xb1*Xw1*Xw2 + Error(id/(Xw1*Xw2)), data=d2))\n```\n\n\nnlme:\n```R\nanova(lme(Y ~ Xb1*Xb2*Xw1, random=~1 | id, method=\"ML\", data=d1))\n```\n\n\n\nlme4:\n```R    \nanova(lmer(Y ~ Xb1*Xw1*Xw2 + (1|id) + (1|Xw1:id) + (1|Xw2:id), data=d2))    \n```"
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "%load_ext rmagic",
     "prompt_number": 23,
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "The rmagic extension is already loaded. To reload it, use:\n  %reload_ext rmagic\n"
      }
     ],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "%R library(lme4)",
     "prompt_number": 24,
     "outputs": [
      {
       "output_type": "display_data",
       "metadata": {},
       "text": "Loading required package: lattice\nLoading required package: Matrix\n"
      }
     ],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "%R library(lmerTest)",
     "prompt_number": 25,
     "outputs": [
      {
       "output_type": "display_data",
       "metadata": {},
       "text": "KernSmooth 2.23 loaded\nCopyright M. P. Wand 1997-2009\n\nAttaching package: ‘lmerTest’\n\nThe following object is masked from ‘package:lme4’:\n\n    lmer\n\nThe following object is masked from ‘package:stats’:\n\n    step\n\n"
      }
     ],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "%%R -i df_oc_vis\nmymod_vis <- lmer(dat ~ AgeGroup*withinORcross*hemi+(1|CCID)+(1|withinORcross:CCID)+(1|hemi:CCID),data=df_oc_vis)    ",
     "prompt_number": 28,
     "outputs": [],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "%%R\nsummary(mymod_vis)",
     "prompt_number": 29,
     "outputs": [
      {
       "output_type": "display_data",
       "metadata": {},
       "text": "Linear mixed model fit by REML ['merModLmerTest']\nFormula: dat ~ AgeGroup * withinORcross * hemi + (1 | CCID) + (1 | withinORcross:CCID) +      (1 | hemi:CCID) \n   Data: df_oc_vis \n\nREML criterion at convergence: -2653.211 \n\nRandom effects:\n Groups             Name        Variance  Std.Dev.\n hemi:CCID          (Intercept) 0.000e+00 0.000000\n withinORcross:CCID (Intercept) 3.824e-04 0.019556\n CCID               (Intercept) 8.594e-05 0.009271\n Residual                       3.194e-03 0.056518\nNumber of obs: 980, groups: hemi:CCID, 248; withinORcross:CCID, 248; CCID, 124\n\nFixed effects:\n                                          Estimate Std. Error         df\n(Intercept)                               0.105123   0.006012 552.000000\nAgeGroupOld                               0.007466   0.009399 550.600000\nAgeGroupYoung                            -0.018507   0.010751 546.200000\nwithinORcrosswithin                      -0.044847   0.008294 312.700000\nhemiR                                    -0.009431   0.007567 723.600000\nAgeGroupOld:withinORcrosswithin           0.003333   0.012974 312.100000\nAgeGroupYoung:withinORcrosswithin         0.021001   0.014868 310.300000\nAgeGroupOld:hemiR                         0.002723   0.011806 722.000000\nAgeGroupYoung:hemiR                       0.034662   0.013421 716.700000\nwithinORcrosswithin:hemiR                 0.012690   0.010599 718.700000\nAgeGroupOld:withinORcrosswithin:hemiR    -0.005706   0.016563 717.800000\nAgeGroupYoung:withinORcrosswithin:hemiR  -0.041305   0.018922 715.100000\n                                        t value Pr(>|t|)    \n(Intercept)                              17.486  < 2e-16 ***\nAgeGroupOld                               0.794   0.4273    \nAgeGroupYoung                            -1.721   0.0857 .  \nwithinORcrosswithin                      -5.407 1.27e-07 ***\nhemiR                                    -1.246   0.2130    \nAgeGroupOld:withinORcrosswithin           0.257   0.7974    \nAgeGroupYoung:withinORcrosswithin         1.412   0.1588    \nAgeGroupOld:hemiR                         0.231   0.8177    \nAgeGroupYoung:hemiR                       2.583   0.0100 *  \nwithinORcrosswithin:hemiR                 1.197   0.2316    \nAgeGroupOld:withinORcrosswithin:hemiR    -0.345   0.7306    \nAgeGroupYoung:withinORcrosswithin:hemiR  -2.183   0.0294 *  \n---\nSignif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1\n\nCorrelation of Fixed Effects:\n            (Intr) AgGrpO AgGrpY wthnOR hemiR  AgGO:OR AgGY:OR AgGO:R AgGY:R\nAgeGroupOld -0.640                                                          \nAgeGroupYng -0.559  0.358                                                   \nwthnORcrssw -0.695  0.445  0.389                                            \nhemiR       -0.617  0.395  0.345  0.447                                     \nAgGrpOld:OR  0.444 -0.695 -0.248 -0.639 -0.286                              \nAgGrpYng:OR  0.388 -0.248 -0.693 -0.558 -0.250  0.357                       \nAgGrpOld:hR  0.396 -0.618 -0.221 -0.287 -0.641  0.448   0.160               \nAgGrpYng:hR  0.348 -0.223 -0.620 -0.252 -0.564  0.161   0.449   0.361       \nwthnORcrs:R  0.441 -0.282 -0.246 -0.633 -0.714  0.404   0.353   0.458  0.403\nAgGrpO:OR:R -0.282  0.440  0.158  0.405  0.457 -0.633  -0.226  -0.713 -0.258\nAgGrpY:OR:R -0.247  0.158  0.440  0.354  0.400 -0.227  -0.634  -0.256 -0.709\n            wtOR:R AGO:OR:\nAgeGroupOld               \nAgeGroupYng               \nwthnORcrssw               \nhemiR                     \nAgGrpOld:OR               \nAgGrpYng:OR               \nAgGrpOld:hR               \nAgGrpYng:hR               \nwthnORcrs:R               \nAgGrpO:OR:R -0.640        \nAgGrpY:OR:R -0.560  0.358 \n"
      }
     ],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "%%R\nanova(mymod_vis)",
     "prompt_number": 30,
     "outputs": [
      {
       "output_type": "display_data",
       "metadata": {},
       "text": "Analysis of Variance Table of type 3  with  Satterthwaite \napproximation for degrees of freedom\n                            Df   Sum Sq  Mean Sq F value  Denom    Pr(>F)    \nAgeGroup                     2 0.011187 0.005594  1.7611 115.49   0.17644    \nwithinORcross                1 0.244147 0.244147 68.5917 118.61 2.101e-13 ***\nhemi                         1 0.000001 0.000001  0.1642 715.78   0.68548    \nAgeGroup:withinORcross       2 0.000010 0.000005  0.0012 118.81   0.99876    \nAgeGroup:hemi                2 0.007993 0.003997  1.2676 716.14   0.28213    \nwithinORcross:hemi           1 0.000262 0.000262  0.1537 715.79   0.69510    \nAgeGroup:withinORcross:hemi  2 0.015924 0.007962  2.4925 716.15   0.08342 .  \n---\nSignif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1\n"
      }
     ],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "%%R\nplot(mymod_vis)",
     "prompt_number": 31,
     "outputs": [
      {
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAeAAAAHgCAIAAADytinCAAAgAElEQVR4nOzdZ1xU19bA4f8MvXcU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\n",
       "metadata": {}
      }
     ],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "anova(lmer(dat ~ AgeGroup*withinORcross*hemi + (1|CCID) + (1|withinORcross:CCID) + (1|hemi:CCID), data=d2))    ",
     "outputs": [],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {},
     "cell_type": "heading",
     "source": "Graveyard",
     "level": 2
    },
    {
     "metadata": {},
     "cell_type": "markdown",
     "source": "BayesFactor package"
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "summary(aov(RT ~ shape * color + Error(ID/(shape * color)), data = puzzles))",
     "outputs": [],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "bf = anovaBF(RT ~ shape*color + ID, data = puzzles, \n             whichRandom=\"ID\")",
     "outputs": [],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "%%R\nsummary(aov(dat ~ AgeGroup * withinORcross * hemi + Error(CCID/(withinORcross * hemi)), data = df_oc))",
     "outputs": [],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "bf = anovaBF(RT ~ shape*color + ID, data = puzzles, \n             whichRandom=\"ID\")",
     "outputs": [],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "mymod_vis <- lmer(dat ~ AgeGroup*withinORcross*hemi+(1|CCID)+(1|withinORcross:CCID)+(1|hemi:CCID),data=df_oc_vis)    ",
     "outputs": [],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "figname = '/tmp/tmp_fig.png'",
     "outputs": [],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "figname = '/tmp/tmp_fig.svg'\n%R -i figname Graph <- plot(bf)\n%R svg(filename=figname, antialias = \"subpixel\", bg=\"transparent\"); plot(Graph); dev.off();\n#%R png(height=1, width=1, res=300, filename=figname, antialias = \"subpixel\", bg=\"transparent\"); plot(Graph); dev.off();",
     "prompt_number": 42,
     "outputs": [
      {
       "png": "iVBORw0KGgoAAAANSUhEUgAAAeAAAAHgCAIAAADytinCAAAgAElEQVR4nO3daVwUZ7o28Luh6Y1u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\n",
       "output_type": "display_data",
       "metadata": {}
      },
      {
       "output_type": "stream",
       "text": "Error in plot.window(...) : need finite 'xlim' values\nIn addition: Warning messages:\n1: In min(x) : no non-missing arguments to min; returning Inf\n2: In max(x) : no non-missing arguments to max; returning -Inf\n3: In min(x) : no non-missing arguments to min; returning Inf\n4: In max(x) : no non-missing arguments to max; returning -Inf\nError in plot.window(...) : need finite 'xlim' values\n",
       "stream": "stdout"
      }
     ],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "figname = '/tmp/tmp_semfig.png'\n%R -i figname Graph <- semPaths(fit,title=FALSE, whatLabels='omit', layoutr=\"spring\", nCharNodes=10, residuals=FALSE); print('Graph written')\n%R png(height=1000, width=2000, res=3000, filename=figname, antialias = \"subpixel\", bg=\"transparent\"); plot(Graph); dev.off();\n[1] \"Graph written\"\n",
     "outputs": [],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "%%R -i df_oc_vis\nbf = anovaBF(dat ~ AgeGroup*withinORcross + CCID, data = df_oc_vis, whichRandom=\"CCID\")\n",
     "prompt_number": 10,
     "outputs": [
      {
       "text": "\r  |                                                        \r  |                                                  |   0%\r  |                                                        \r  |++++++++++                                        |  20%\r  |                                                        \r  |++++++++++++++++++++                              |  40%\r  |                                                        \r  |++++++++++++++++++++++++++++++                    |  60%\r  |                                                        \r  |++++++++++++++++++++++++++++++++++++++++          |  80%\r  |                                                        \r  |++++++++++++++++++++++++++++++++++++++++++++++++++| 100%\n",
       "output_type": "display_data",
       "metadata": {}
      }
     ],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "%R plot(bf)",
     "prompt_number": 29,
     "outputs": [
      {
       "png": "iVBORw0KGgoAAAANSUhEUgAAAeAAAAHgCAIAAADytinCAAAgAElEQVR4nO3daVwUZ7o28Luh6Y1u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\n",
       "output_type": "display_data",
       "metadata": {}
      }
     ],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "",
     "prompt_number": 26,
     "outputs": [
      {
       "output_type": "stream",
       "text": "Object `plot` not found.\n",
       "stream": "stdout"
      }
     ],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "%R plot",
     "outputs": [],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "",
     "outputs": [],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "%%R\nbfWithoutID = lmBF(dat ~ AgeGroup * withinORcross, data = df_oc_vis)\nbfWithoutID",
     "prompt_number": 12,
     "outputs": [
      {
       "text": "\r  |                                                                            \r  |                                                                      |   0%\r  |                                                                            \r  |=                                                                     |   1%\r  |                                                                            \r  |=                                                                     |   2%\r  |                                                                            \r  |==                                                                    |   3%\r  |                                                                            \r  |===                                                                   |   4%\r  |                                                                            \r  |====                                                                  |   5%\r  |                                                                            \r  |====                                                                  |   6%\r  |                                                                            \r  |=====                                                                 |   7%\r  |                                                                            \r  |======                                                                |   8%\r  |                                                                            \r  |======                                                                |   9%\r  |                                                                            \r  |=======                                                               |  10%\r  |                                                                            \r  |========                                                              |  11%\r  |                                                                            \r  |========                                                              |  12%\r  |                                                                            \r  |=========                                                             |  13%\r  |                                                                            \r  |==========                                                            |  14%\r  |                                                                            \r  |==========                                                            |  15%\r  |                                                                            \r  |===========                                                           |  16%\r  |                                                                            \r  |============                                                          |  17%\r  |                                                                            \r  |=============                                                         |  18%\r  |                                                                            \r  |=============                                                         |  19%\r  |                                                                            \r  |==============                                                        |  20%\r  |                                                                            \r  |===============                                                       |  21%\r  |                                                                            \r  |===============                                                       |  22%\r  |                                                                            \r  |================                                                      |  23%\r  |                                                                            \r  |=================                                                     |  24%\r  |                                                                            \r  |==================                                                    |  25%\r  |                                                                            \r  |==================                                                    |  26%\r  |                                                                            \r  |===================                                                   |  27%\r  |                                                                            \r  |====================                                                  |  28%\r  |                                                                            \r  |====================                                                  |  29%\r  |                                                                            \r  |=====================                                                 |  30%\r  |                                                                            \r  |======================                                                |  31%\r  |                                                                            \r  |======================                                                |  32%\r  |                                                                            \r  |=======================                                               |  33%\r  |                                                                            \r  |========================                                              |  34%\r  |                                                                            \r  |========================                                              |  35%\r  |                                                                            \r  |=========================                                             |  36%\r  |                                                                            \r  |==========================                                            |  37%\r  |                                                                            \r  |===========================                                           |  38%\r  |                                                                            \r  |===========================                                           |  39%\r  |                                                                            \r  |============================                                          |  40%\r  |                                                                            \r  |=============================                                         |  41%\r  |                                                                            \r  |=============================                                         |  42%\r  |                                                                            \r  |==============================                                        |  43%\r  |                                                                            \r  |===============================                                       |  44%\r  |                                                                            \r  |================================                                      |  45%\r  |                                                                            \r  |================================                                      |  46%\r  |                                                                            \r  |=================================                                     |  47%\r  |                                                                            \r  |==================================                                    |  48%\r  |                                                                            \r  |==================================                                    |  49%\r  |                                                                            \r  |===================================                                   |  50%\r  |                                                                            \r  |====================================                                  |  51%\r  |                                                                            \r  |====================================                                  |  52%\r  |                                                                            \r  |=====================================                                 |  53%\r  |                                                                            \r  |======================================                                |  54%\r  |                                                                            \r  |======================================                                |  55%\r  |                                                                            \r  |=======================================                               |  56%\r  |                                                                            \r  |========================================                              |  57%\r  |                                                                            \r  |=========================================                             |  58%\r  |                                                                            \r  |=========================================                             |  59%\r  |                                                                            \r  |==========================================                            |  60%\r  |                                                                            \r  |===========================================                           |  61%\r  |                                                                            \r  |===========================================                           |  62%\r  |                                                                            \r  |============================================                          |  63%\r  |                                                                            \r  |=============================================                         |  64%\r  |                                                                            \r  |==============================================                        |  65%\r  |                                                                            \r  |==============================================                        |  66%\r  |                                                                            \r  |===============================================                       |  67%\r  |                                                                            \r  |================================================                      |  68%\r  |                                                                            \r  |================================================                      |  69%\r  |                                                                            \r  |=================================================                     |  70%\r  |                                                                            \r  |==================================================                    |  71%\r  |                                                                            \r  |==================================================                    |  72%\r  |                                                                            \r  |===================================================                   |  73%\r  |                                                                            \r  |====================================================                  |  74%\r  |                                                                            \r  |====================================================                  |  75%\r  |                                                                            \r  |=====================================================                 |  76%\r  |                                                                            \r  |======================================================                |  77%\r  |                                                                            \r  |=======================================================               |  78%\r  |                                                                            \r  |=======================================================               |  79%\r  |                                                                            \r  |========================================================              |  80%\r  |                                                                            \r  |=========================================================             |  81%\r  |                                                                            \r  |=========================================================             |  82%\r  |                                                                            \r  |==========================================================            |  83%\r  |                                                                            \r  |===========================================================           |  84%\r  |                                                                            \r  |============================================================          |  85%\r  |                                                                            \r  |============================================================          |  86%\r  |                                                                            \r  |=============================================================         |  87%\r  |                                                                            \r  |==============================================================        |  88%\r  |                                                                            \r  |==============================================================        |  89%\r  |                                                                            \r  |===============================================================       |  90%\r  |                                                                            \r  |================================================================      |  91%\r  |                                                                            \r  |================================================================      |  92%\r  |                                                                            \r  |=================================================================     |  93%\r  |                                                                            \r  |==================================================================    |  94%\r  |                                                                            \r  |==================================================================    |  95%\r  |                                                                            \r  |===================================================================   |  96%\r  |                                                                            \r  |====================================================================  |  97%\r  |                                                                            \r  |===================================================================== |  98%\r  |                                                                            \r  |===================================================================== |  99%\r  |                                                                            \r  |======================================================================| 100%\nBayes factor analysis\n--------------\n[1] AgeGroup * withinORcross : 1.537342e+16 ±4.29%\n\nAgainst denominator:\n  Intercept only \n---\nBayes factor type: BFlinearModel, JZS\n\n",
       "output_type": "display_data",
       "metadata": {}
      }
     ],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "%%R\nbfOnlyID = lmBF(dat ~ CCID, whichRandom = \"CCID\", data = df_oc_vis)\nbf2 = bfWithoutID/bfOnlyID\nbf2",
     "prompt_number": 13,
     "outputs": [
      {
       "text": "Bayes factor analysis\n--------------\n[1] AgeGroup * withinORcross : 1.045082e+18 ±4.29%\n\nAgainst denominator:\n  dat ~ CCID \n---\nBayes factor type: BFlinearModel, JZS\n\n",
       "output_type": "display_data",
       "metadata": {}
      }
     ],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "%%R \nbfall = c(bf, bf2)\nbfall",
     "prompt_number": 23,
     "outputs": [
      {
       "text": "Bayes factor analysis\n--------------\n[1] withinORcross + CCID                                     : 4.238653e+20 ±0.97%\n[2] AgeGroup + CCID                                          : 0.08356938   ±0.73%\n[3] withinORcross + AgeGroup + CCID                          : 4.212518e+19 ±2.37%\n[4] withinORcross + AgeGroup + withinORcross:AgeGroup + CCID : 9.575881e+17 ±1.36%\n[5] AgeGroup * withinORcross                                 : 1.045082e+18 ±4.29%\n\nAgainst denominator:\n  dat ~ CCID \n---\nBayes factor type: BFlinearModel, JZS\n\n",
       "output_type": "display_data",
       "metadata": {}
      }
     ],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "%%R\nres = bf[4]/bf2\nres",
     "prompt_number": 24,
     "outputs": [
      {
       "text": "Bayes factor analysis\n--------------\n[1] withinORcross + AgeGroup + withinORcross:AgeGroup + CCID : 0.9162804 ±4.5%\n\nAgainst denominator:\n  dat ~ AgeGroup * withinORcross \n---\nBayes factor type: BFlinearModel, JZS\n\n",
       "output_type": "display_data",
       "metadata": {}
      }
     ],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "%%R\n?plot",
     "prompt_number": 27,
     "outputs": [
      {
       "output_type": "stream",
       "text": "R Help on ‘plot’plot                 package:graphics                  R Documentation\n\n_\bG_\be_\bn_\be_\br_\bi_\bc _\bX-_\bY _\bP_\bl_\bo_\bt_\bt_\bi_\bn_\bg\n\n_\bD_\be_\bs_\bc_\br_\bi_\bp_\bt_\bi_\bo_\bn:\n\n     Generic function for plotting of R objects.  For more details\n     about the graphical parameter arguments, see ‘par’.\n\n     For simple scatter plots, ‘plot.default’ will be used.  However,\n     there are ‘plot’ methods for many R objects, including\n     ‘function’s, ‘data.frame’s, ‘density’ objects, etc.  Use\n     ‘methods(plot)’ and the documentation for these.\n\n_\bU_\bs_\ba_\bg_\be:\n\n     plot(x, y, ...)\n     \n_\bA_\br_\bg_\bu_\bm_\be_\bn_\bt_\bs:\n\n       x: the coordinates of points in the plot. Alternatively, a\n          single plotting structure, function or _any R object with a\n          ‘plot’ method_ can be provided.\n\n       y: the y coordinates of points in the plot, _optional_ if ‘x’ is\n          an appropriate structure.\n\n     ...: Arguments to be passed to methods, such as graphical\n          parameters (see ‘par’).  Many methods will accept the\n          following arguments:\n\n          ‘type’ what type of plot should be drawn.  Possible types are\n\n                • ‘\"p\"’ for *p*oints,\n\n                • ‘\"l\"’ for *l*ines,\n\n                • ‘\"b\"’ for *b*oth,\n\n                • ‘\"c\"’ for the lines part alone of ‘\"b\"’,\n\n                • ‘\"o\"’ for both ‘*o*verplotted’,\n\n                • ‘\"h\"’ for ‘*h*istogram’ like (or ‘high-density’)\n                  vertical lines,\n\n                • ‘\"s\"’ for stair *s*teps,\n\n                • ‘\"S\"’ for other *s*teps, see ‘Details’ below,\n\n                • ‘\"n\"’ for no plotting.\n\n              All other ‘type’s give a warning or an error; using,\n              e.g., ‘type = \"punkte\"’ being equivalent to ‘type = \"p\"’\n              for S compatibility.  Note that some methods, e.g.\n              ‘plot.factor’, do not accept this.\n\n          ‘main’ an overall title for the plot: see ‘title’.\n\n          ‘sub’ a sub title for the plot: see ‘title’.\n\n          ‘xlab’ a title for the x axis: see ‘title’.\n\n          ‘ylab’ a title for the y axis: see ‘title’.\n\n          ‘asp’ the y/x aspect ratio, see ‘plot.window’.\n\n_\bD_\be_\bt_\ba_\bi_\bl_\bs:\n\n     The two step types differ in their x-y preference: Going from\n     (x1,y1) to (x2,y2) with x1 < x2, ‘type = \"s\"’ moves first\n     horizontal, then vertical, whereas ‘type = \"S\"’ moves the other\n     way around.\n\n_\bS_\be_\be _\bA_\bl_\bs_\bo:\n\n     ‘plot.default’, ‘plot.formula’ and other methods; ‘points’,\n     ‘lines’, ‘par’.\n\n     For X-Y-Z plotting see ‘contour’, ‘persp’ and ‘image’.\n\n_\bE_\bx_\ba_\bm_\bp_\bl_\be_\bs:\n\n     require(stats)\n     plot(cars)\n     lines(lowess(cars))\n     \n     plot(sin, -pi, 2*pi) # see ?plot.function\n     \n     ## Discrete Distribution Plot:\n     plot(table(rpois(100, 5)), type = \"h\", col = \"red\", lwd = 10,\n          main = \"rpois(100, lambda = 5)\")\n     \n     ## Simple quantiles/ECDF, see ecdf() {library(stats)} for a better one:\n     plot(x <- sort(rnorm(47)), type = \"s\", main = \"plot(x, type = \\\"s\\\")\")\n     points(x, cex = .5, col = \"dark red\")\n     \n",
       "stream": "stdout"
      }
     ],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "",
     "outputs": [],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "http://www.uni-kiel.de/psychologie/rexrepos/posts/anovaMixed.html#three-way-split-plot-factorial-anova-spf-pq-cdot-r-design\n    \n    Three-way split-plot-factorial ANOVA (SPF-pq⋅r design)\nConventional analysis using aov()\n\nsummary(aov(Y ~ Xb1*Xb2*Xw1 + Error(id/Xw1), data=d1))\n\nanova(lmer(Y ~ Xb1*Xw1*Xw2 + (1|id) + (1|Xw1:id) + (1|Xw2:id), data=d2))\n\n\nanova(lmer(Y ~ Xb1*Xb2*Xw1 + (1|id), data=d1))",
     "outputs": [],
     "language": "python",
     "collapsed": false
    },
    {
     "metadata": {
      "run_control": {
       "breakpoint": false
      }
     },
     "cell_type": "code",
     "input": "",
     "outputs": [],
     "language": "python",
     "collapsed": false
    }
   ],
   "metadata": {}
  }
 ],
 "metadata": {
  "gist_id": "8946387",
  "name": "",
  "css": [
   ""
  ]
 },
 "nbformat": 3
}