liebke

(use '(incanter core stats charts datasets))
;; load the iris dataset
(def iris (to-matrix (get-dataset :iris)))
;; run the pca on the first four columns only
(def pca (principal-components (sel iris :cols (range 4))))
;; extract the first two principal components
(def pc1 (sel (:rotation pca) :cols 0))
(def pc2 (sel (:rotation pca) :cols 1))

;; project the first four dimension of the iris data onto the first

liebke

(use '(incanter core stats charts))

;; Overlay a xy-plot on a histogram 

;; Plot a histogram of a normal sample of size 1000, using the :density option
;; so that the y-axis represents density instead of the default frequency.
;; Then plot a pdf curve for the normal distribution on the same chart.

(def x (range -3 3 0.01))
(doto (histogram (sample-normal 1000) :density true)

liebke

(use '(incanter core charts))

;; plot the sine and cosine functions with
;; the function-plot and add-function functions

(doto (function-plot sin (- Math/PI) Math/PI)
      (add-function cos (- Math/PI) Math/PI)
      view)

liebke

;; plot a user defined function and its derivative
(use '(incanter core charts optimize))

;; define the function, x^3 + 2x^2 + 2x + 3
(defn cubic [x] (+ (* x x x) (* 2 x x) (* 2 x) 3))

;; use the derivative function to get a function
;; that approximates the derivative of cubic
(def deriv-cubic (derivative cubic))

liebke

;; example of performing PCA on Fisher's iris data

(use '(incanter core stats charts datasets))

(def iris (to-matrix (get-dataset :iris)))
(view iris)

(def X (sel iris :cols (range 4)))
(def species (sel iris :cols 4))
(def pca (principal-components X))

liebke

;; example of testing the significance of a correlation
;; value with a permutation test 

(use '(incanter core stats charts datasets))

;; load the data
(def data (to-matrix (get-dataset :us-arrests)))
(def assault (sel data :cols 2))
(def urban-pop (sel data :cols 3))

liebke

;; compare the means of different treatments in the plant-growth data set

(use '(incanter core stats charts datasets))

;; load the plant-growth data
(def plant-growth (to-matrix (get-dataset :plant-growth)))

;; create box-plots of the three treatment groups
(view (box-plot (sel plant-growth :cols 0) 
                :group-by (sel plant-growth :cols 1)))

liebke

;; load the necessary libraries
(use '(incanter core stats charts datasets))

;; load the NIST filip data set
;; see information on this data set at http://www.itl.nist.gov/div898/strd/lls/data/Filip.shtml
(def data (to-matrix (get-dataset :filip)))

(def y (sel data :cols 0))

;; use the sweep function to center the x values to reduce collinearity of the polynomial terms

liebke

(use '(incanter core optimize datasets charts))

;; Chwirut data set from NIST
;; http://www.itl.nist.gov/div898/strd/nls/data/LINKS/DATA/Chwirut1.dat
(def data (to-matrix (get-dataset :chwirut)))
(def x (sel data :cols 1))
(def y (sel data :cols 0))

;; define model function: y = exp(-b1*x)/(b2+b3*x) + e
(defn f [theta x]

liebke

(use '(incanter core charts))

(def x (range (* -2 Math/PI) (* 2 Math/PI) 0.01))
(def plot (xy-plot x (sin x)))
(view plot)

;; annotate the plot
(doto plot
  (add-pointer (- Math/PI) (sin (- Math/PI)) 
               :text "(-pi, (sin -pi))")