ISLR - Learning Classification using the Stock Market Data

1. Running Logistic Regression models on the Stock Market dataset

summary(Smarket)

      Year           Lag1                Lag2                Lag3                Lag4                Lag5         
 Min.   :2001   Min.   :-4.922000   Min.   :-4.922000   Min.   :-4.922000   Min.   :-4.922000   Min.   :-4.92200  
 1st Qu.:2002   1st Qu.:-0.639500   1st Qu.:-0.639500   1st Qu.:-0.640000   1st Qu.:-0.640000   1st Qu.:-0.64000  
 Median :2003   Median : 0.039000   Median : 0.039000   Median : 0.038500   Median : 0.038500   Median : 0.03850  
 Mean   :2003   Mean   : 0.003834   Mean   : 0.003919   Mean   : 0.001716   Mean   : 0.001636   Mean   : 0.00561  
 3rd Qu.:2004   3rd Qu.: 0.596750   3rd Qu.: 0.596750   3rd Qu.: 0.596750   3rd Qu.: 0.596750   3rd Qu.: 0.59700  
 Max.   :2005   Max.   : 5.733000   Max.   : 5.733000   Max.   : 5.733000   Max.   : 5.733000   Max.   : 5.73300  
     Volume           Today           Direction 
 Min.   :0.3561   Min.   :-4.922000   Down:602  
 1st Qu.:1.2574   1st Qu.:-0.639500   Up  :648  
 Median :1.4229   Median : 0.038500             
 Mean   :1.4783   Mean   : 0.003138             
 3rd Qu.:1.6417   3rd Qu.: 0.596750             
 Max.   :3.1525   Max.   : 5.733000             

• Running a logistic regression model

glm_probs[1:5]

        1         2         3         4         5 
0.5070841 0.4814679 0.4811388 0.5152224 0.5107812 

Checking our performance

#Getting our classification performance
mean(glm_pred==Smarket$Direction)

[1] 0.5216

• Splitting into train-test

Checking performance on test data

mean(glm_pred_test == Smarket_Direction_2005)

[1] 0.4801587

• Classification using only lag1 and lag2 terms

glm_fit_train2 <- glm(Direction ~ Lag1 + Lag2, data = Smarket, family = binomial, subset = train)
glm_probs_test2 <- predict(glm_fit_train2, newdata = Smarket[!train,], type = "response")
glm_pred_test2 <- ifelse(glm_probs_test2>0.5, "Up", "Down")

Checking performance of updated classification codes

table(glm_pred_test2, Smarket_Direction_2005)

              Smarket_Direction_2005
glm_pred_test2 Down  Up
          Down   35  35
          Up     76 106

mean(glm_pred_test2 == Smarket_Direction_2005)

[1] 0.5595238

---

2. Running Linear Discriminant Analysis on the Stock Market dataset

rm(ls=list())

Error in rm(ls = list()) : ... must contain names or character strings

Fitting a LDA model

Checking what the performance is on test data (year = 2005)

mean(lda_pred$class==Smarket_2005$Direction)

[1] 0.5595238

---

3. K-Nearest Neighbours classification

mean(knn_pred == Smarket$Direction[!train])

[1] 0.5

K-Nearest Neighbors

library(class) ?knn attach(Smarket) Xlag=cbind(Lag1,Lag2) train=Year<2005 knn.pred=knn(Xlag[train,],Xlag[!train,],Direction[train],k=1) table(knn.pred,Direction[!train]) mean(knn.pred==Direction[!train])