Solution:
Command Sequence
> data<-read.csv(file.choose(),header=T)
> close<-data$Close[10:95]
> close.ts<-ts(close,deltat=1/252)
> close.ts
Time Series:
Start = c(1, 1)
End = c(1, 86)
Frequency = 252
[1] 5380.35 5362.95 5366.30 5421.00 5412.85 5415.35 5386.70 5350.25 5334.60
[10] 5287.80 5315.05 5258.50 5253.75 5274.00 5225.70 5238.40 5342.10 5358.70
[19] 5363.45 5390.00 5431.00 5435.35 5577.65 5610.00 5600.05 5554.25 5691.15
[28] 5669.60 5673.90 5663.45 5649.50 5703.30 5718.80 5731.25 5787.60 5746.95
[37] 5676.00 5704.60 5652.15 5708.05 5676.05 5687.25 5648.00 5660.25 5718.70
[46] 5684.25 5717.15 5691.40 5705.30 5664.30 5665.60 5597.90 5619.70 5645.05
[55] 5697.70 5704.20 5724.40 5760.10 5738.75 5686.25 5683.70 5666.95 5631.00
[64] 5574.05 5571.40 5571.55 5614.80 5627.75 5626.60 5635.90 5727.45 5825.00
[73] 5879.85 5870.95 5889.25 5900.50 5930.90 5907.40 5908.90 5898.80 5888.00
[82] 5851.50 5879.60 5857.90 5896.80 5929.60
> summary(close.ts)
Min. 1st Qu. Median Mean 3rd Qu. Max.
5226 5465 5665 5624 5723 5931
> close.diff<-diff(close.ts)
> close.diff
Time Series:
Start = c(1, 2)
End = c(1, 86)
Frequency = 252
[1] -17.40 3.35 54.70 -8.15 2.50 -28.65 -36.45 -15.65 -46.80 27.25
[11] -56.55 -4.75 20.25 -48.30 12.70 103.70 16.60 4.75 26.55 41.00
[21] 4.35 142.30 32.35 -9.95 -45.80 136.90 -21.55 4.30 -10.45 -13.95
[31] 53.80 15.50 12.45 56.35 -40.65 -70.95 28.60 -52.45 55.90 -32.00
[41] 11.20 -39.25 12.25 58.45 -34.45 32.90 -25.75 13.90 -41.00 1.30
[51] -67.70 21.80 25.35 52.65 6.50 20.20 35.70 -21.35 -52.50 -2.55
[61] -16.75 -35.95 -56.95 -2.65 0.15 43.25 12.95 -1.15 9.30 91.55
[71] 97.55 54.85 -8.90 18.30 11.25 30.40 -23.50 1.50 -10.10 -10.80
[81] -36.50 28.10 -21.70 38.90 32.80
> return<-data.diff/lag(close.ts,k=-1)
Error: object 'data.diff' not found
> return<-close.diff/lag(close.ts,k=-1)
> plot(return,main="Final Returns from 10 th to 95th day")
> plot(return,main="Final Returns from 10th to 95th day")
Assignment -2: 1-700 data is available, Predict the data from 701-850, use the GLM estimation using LOGIT Analysis for the same.
Solution:
Command Sequence
> z<-read.csv(file.choose(),header=T)
> head(z)
age ed employ address income debtinc creddebt othdebt default
1 41 3 17 12 176 9.3 11.36 5.01 1
2 27 1 10 6 31 17.3 1.36 4.00 0
3 40 1 15 14 55 5.5 0.86 2.17 0
4 41 1 15 14 120 2.9 2.66 0.82 0
5 24 2 2 0 28 17.3 1.79 3.06 1
6 41 2 5 5 25 10.2 0.39 2.16 0
> z.data<-z[1:700,1:9]
> sapply(z.data,mean)
age ed employ address income debtinc creddebt
34.8600000 1.7228571 8.3885714 8.2785714 45.6014286 10.2605714 1.5534571
othdebt default
3.0582286 0.2614286
> z.data$ed<-factor(z.data$ed)
> logit.est<-glm(default~age+employ+address+income+debtinc+creddebt+othdebt,data=z.data,family="binomial")
> logitdata.est<-glm(default~age+employ+address+income+debtinc+creddebt+othdebt,data=z.data,family="binomial")
> summary(logitdata.est)
Call:
glm(formula = default ~ age + employ + address + income + debtinc +
creddebt + othdebt, family = "binomial", data = z.data)
Deviance Residuals:
Min 1Q Median 3Q Max
-2.3659 -0.6516 -0.2882 0.2625 2.9757
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.376573 0.571560 -2.408 0.0160 *
age 0.033712 0.017342 1.944 0.0519 .
employ -0.265086 0.031999 -8.284 < 2e-16 ***
address -0.103960 0.023192 -4.483 7.38e-06 ***
income -0.007566 0.008095 -0.935 0.3500
debtinc 0.065099 0.030621 2.126 0.0335 *
creddebt 0.628475 0.113759 5.525 3.30e-08 ***
othdebt 0.070761 0.077682 0.911 0.3623
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 804.36 on 699 degrees of freedom
Residual deviance: 552.20 on 692 degrees of freedom
AIC: 568.2
Number of Fisher Scoring iterations: 6
> confint.default(logitdata.est)
2.5 % 97.5 %
(Intercept) -2.4968094058 -0.25633710
age -0.0002768342 0.06770153
employ -0.3278025120 -0.20236959
address -0.1494167558 -0.05850405
income -0.0234310297 0.00829931
debtinc 0.0050836856 0.12511410
creddebt 0.4055112145 0.85143903
othdebt -0.0814939858 0.22301505
> logitdata.eg<-with(z[701:850,1:8],data.frame(age=age,employ=employ,address=address,income=income,debtinc=debtinc,creddebt=creddebt,othdebt=othdebt,ed=factor(1:3)))
> logitdata.eg$prob<-predict(logitdata.est,newdata=logitdata.eg,type="response")
> head(logitdata.eg)
age employ address income debtinc creddebt othdebt ed prob
1 36 16 13 32 10.9 0.54 2.94 1 0.00865515
2 50 6 27 21 12.9 1.32 1.39 2 0.07730314
3 40 9 9 33 17.0 4.88 0.73 3 0.65161010
4 31 5 7 23 2.0 0.05 0.41 1 0.08564367
5 29 4 0 24 7.8 0.87 1.01 2 0.37406261
6 25 1 3 14 9.9 0.23 1.15 3 0.41429185
No comments:
Post a Comment