Scribd
Upload
18 views2 pages

Regression Analysis of Business Returns

The document outlines an R program for analyzing the relationship between business returns and two predictors: the number of branches and the age of the business. It includes steps for reading data, fitting linear and nonlinear regression models, estimating parameters, and determining p-values for the estimated parameters. The program utilizes the lm function for linear regression and the nlsML function for nonlinear regression.

Uploaded by

mjnderi7
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
18 views2 pages

Regression Analysis of Business Returns

The document outlines an R program for analyzing the relationship between business returns and two predictors: the number of branches and the age of the business. It includes steps for reading data, fitting linear and nonlinear regression models, estimating parameters, and determining p-values for the estimated parameters. The program utilizes the lm function for linear regression and the nlsML function for nonlinear regression.

Uploaded by

mjnderi7
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

5.

Statistician is interested in regressing the returns (Y) of business on the number of branches
(X1) and the age of business X2. The statistician collected the relevant data form 10 big
companies as shown in the table below

Y 61 47 75 63 79 75 67 47 71 84

X1 19 19 10 17 19 12 18 10 15 15

X2 11 4 8 12 5 12 6 14 13 11

Write a well commented R program that does the following

(i) Reads in data [3 marks]


#creating a data frame

Y<-c(61,47,75,63,79,75,67,47,71,84)

X1<-c(19,19,10,17,19,12,18,10,15,15)

X2<- c(11,4,8,12,5,12,6,14,13,11)

Data<- [Link](Y,X1,X2)

(ii) Estimates the parameters β1 and β2 in the regression model Y = β0 + β1x1 + β2x2 +
e where e ∼ N (0,σ2) [3 marks]
#Fitting a linear regression model

model1<- lm(Y~X1+X2, data= Data)

beta0<- coef(model1)[1]

#Estimates of parameters β1 and β2

beta1<- coef(model1)[2]

beta2<- coef(model1)[3]

#print(model1$coefficients)

(iii) Estimates the parameters a, b and c in the regression model Y = a + b ∗ x1 + exp(c ∗


x2) + e where e ∼ N (0,σ2) [3 marks]
#Library in use

library([Link])

# Fitting a nonlinear model

model2<-nlsML(Y~a+b*X1+exp(c*X2),data=Data,start=list(a=1,b=1,c=1))

# Extraction of the parameter estimates

a<- coef(model2)[1]

b<- coef(model2)[2]

c<- coef(model2)[3]
#print(model2$coefficients)

#Display estimates a, b and c

cat("Estimated a, b and c\n","a = ",a, "b = ",b,"c = ",c)

(iv) Determine the p-values of the estimated parameters in the part (ii) and (iii) above [3
marks]

#p-values of the estimated parameters in the part (ii)

s_model1<-summary(model1)

s_model1

P_values_model1<- s_model1$coefficients[,4]

P_values_model1

#p-values of the estimated parameters in the part (iii)

s_model2<- summary(model2)

s_model2

P_values_model2<- s_model2$coefficients[,4]

P_values_model2

You might also like