gistfile1.txt

Call:
glm(formula = PredictValue ~ HI + BR + Vol + mday + wday, family = binomial(link = "logit"), 
    data = train, na.action = "na.exclude")

Deviance Residuals: 
    Min       1Q   Median       3Q      Max  
-1.7719  -1.1056  -0.7567   1.1626   1.7096  

Coefficients:
             Estimate Std. Error z value Pr(>|z|)   
(Intercept) -12.91453  324.74416  -0.040  0.96828   
HITRUE        0.45297    0.27386   1.654  0.09813
BR1          -0.27223    0.13269  -2.052  0.04021 * - brent, (close - open)
Vol           1.09640    0.84043   1.305  0.19204   
mday2         0.84098    0.54471   1.544  0.12261   
mday3         0.36857    0.53584   0.688  0.49156   
mday4         0.31662    0.54887   0.577  0.56404   
mday5         0.46437    0.53104   0.874  0.38187   
mday6         1.43370    0.55422   2.587  0.00968 ** - 6 день месяца
mday7         0.84815    0.54731   1.550  0.12122   
mday8         0.44385    0.54494   0.814  0.41536   
mday9         0.19402    0.54220   0.358  0.72047   
mday10        0.31292    0.52787   0.593  0.55331   
mday11        0.62069    0.53426   1.162  0.24533   
mday12        0.10654    0.54815   0.194  0.84589   
mday13        1.47491    0.55280   2.668  0.00763 ** - 13 день месяца
mday14        0.46868    0.53182   0.881  0.37817   
mday15        1.23177    0.55128   2.234  0.02546 * 
mday16        0.75166    0.53050   1.417  0.15651   
mday17        0.52328    0.52849   0.990  0.32211   
mday18       -0.27044    0.54778  -0.494  0.62151   
mday19        0.26033    0.53083   0.490  0.62384   
mday20        0.60638    0.52809   1.148  0.25087   
mday21        0.74002    0.53715   1.378  0.16830   
mday22        0.21478    0.53953   0.398  0.69057   
mday23        0.26690    0.53500   0.499  0.61787   
mday24        0.47653    0.52774   0.903  0.36655   
mday25        0.18128    0.53570   0.338  0.73505   
mday26       -0.06555    0.53820  -0.122  0.90306   
mday27       -0.19334    0.54010  -0.358  0.72036   
mday28        0.34553    0.53205   0.649  0.51606   
mday29        0.42750    0.54241   0.788  0.43061   
mday30        0.16579    0.54032   0.307  0.75896   
mday31        0.43723    0.64707   0.676  0.49922   
wday1        12.68585  324.74393   0.039  0.96884   
wday2        12.63665  324.74392   0.039  0.96896   
wday3        12.25958  324.74391   0.038  0.96989   
wday4        12.59046  324.74392   0.039  0.96907   
wday5        12.42069  324.74391   0.038  0.96949   
wday6        11.73706  324.74600   0.036  0.97117   
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

(Dispersion parameter for binomial family taken to be 1)

Null deviance: 1431.6  on 1032  degrees of freedom
Residual deviance: 1370.6  on  993  degrees of freedom
AIC: 1450.6

Number of Fisher Scoring iterations: 1


```{r}

result <- predict(model, train, type = "response", na.action = "na.exclude")
result <- ifelse(result > 0.5, 1, 0)

result <- as.integer(result != train$PredictValue)
1 - mean(result[!is.na(result)])

result <- predict(model, test, type = "response", na.action = "na.exclude")
result <- ifelse(result > 0.5, 1, 0)

result <- as.integer(result !=test$PredictValue)
1 - mean(result[!is.na(result)])


train: [1] 0.6108422
test: [1] 0.5391304