Great Deal! Get Instant $10 FREE in Account on First Order + 10% Cashback on Every Order Order Now

Econ XXXXXXXXXX: Topics in Applied Econometrics Assignment/Report Due:Week 10, 7AM, Monday 12th October 2020 Instructions: • Write your assignment using R Markdown and submit it along with your...

1 answer below »
Econ XXXXXXXXXX: Topics in Applied Econometrics
Assignment/Report
Due:Week 10, 7AM, Monday 12th October 2020
Instructions:
• Write your assignment using R Markdown and submit it along with your assignment
submission file. The assignment file must be in PDF format.
• Submit both your files through the assignment submission link on iLearn.
• Note: Late assignments will be accepted up to 96 hours after the due date and time, but there
will be a deduction of 10% of the total awarded marks for each 24 hour period late or part
thereof that the submission is late (for example, 25 hours late in submission – 20% penalty).
This penalty does not apply for cases in which an application for Special Consideration is made
and approved.
• All questions ca
y equal marks.
Assignment Questions
A survey was conducted in 1977 and 1989 to evaluate " A working mother can establish just as warm
and secure of relationship with her child as a mother who does not work". 2293 respondents
participated in this survey and stored in a file name "gsskidvalue.csv" and the responses were
ecorded as:
1 = strongly disagree (SD), 2 = disagree (D), 3 = agree (A), and 4 = strongly agree (SA).
Data for following variables were also recorded from the respondents.
opinion: respondents opinion towards the survey statement (SD, D, A, SA)
yr89: survey year 1989.
male: male = 1, 0 for female.
white: white = 1, 0 for non-white.
age: age in years.
ed: years of education.
1
prst: occupational prestige.
(a) Provide a summary of your data.
(b) Obtain a summary table that shows sample proportion between the categories opinionand
male. Comment on your results.
(c) Estimate the ordered logit and probit models with dependent variable opinion as a function of
male (no intercept) and provide your output.
(d) Is the estimates of male significant in both models? Test at 1% significance level. (You don’t
need to ca
y out a hypothesis test )
(e) Using both models, calculate the probability for each of the alternatives for male=1 and
male=0 and comment on your results.
(f) Estimate ordered logit and probit models with the dependent variable opinion as a function of
explanatory variables, male, white, age, ed and prst. Provide summary results for both
models.
(g) Are all coefficients in both models individually significant at 5% significance level? comment.
(h) Use the likelihood ratio (LR) test and test the overall significance of both models and comment
on your results.
(i) Obtain odds ratios for the logit model and comment.
(j) Use both models to predict the outcome (opinion) for a male white person when the rest of the
variables are at their mean values. Do you find any significant differences in the predicted
values between the two models?
(k) Make a prediction for the 2292th observation in our sample data which has the following
information : male=0 , white=1 , age=55 , ed=14 and prst=36 . Compare your results
with the actual response and comment.
2
(l) Obtain marginal effects for each of the variables and interpret its values for the variable age.
Comment on your results.
(m) How similar are both models in terms of making predictions for ordinal outcome? would you
prefer one model over the other? why or why not?
END
3
Answered Same Day Oct 07, 2021 Macquarie University

Solution

Pooja answered on Oct 08 2021
147 Votes
a)
gsskidvalue <- read.csv("C:/Users/HP/Desktop/gsskidvalue.csv", header=TRUE)
View(gsskidvalue)
summary(gsskidvalue)
opinion yr83 male white age ed prst
A :856 Min. :1977 Min. :0.0000 Min. :0.0000 Min. :18.00 Min. : 0.00 Min. :12.00
D :723 1st Qu.:1977 1st Qu.:0.0000 1st Qu.:0.0000 1st Qu.:31.00 1st Qu.:11.00 1st Qu.:30.00
SA:417 Median :1977 Median :0.0000 Median :0.0000 Median :42.00 Median :12.00 Median :37.00
SD:297 Mean :1982 Mean :0.4649 Mean :0.1234 Mean :44.94 Mean :12.22 Mean :39.59
3rd Qu.:1989 3rd Qu.:1.0000 3rd Qu.:0.0000 3rd Qu.:58.00 3rd Qu.:14.00 3rd Qu.:50.00
Max. :1989 Max. :1.0000 Max. :1.0000 Max. :89.00 Max. :20.00 Max. :82.00
)
readgender <- table(gsskidvalue$opinion,gsskidvalue$male)
prop.table(readgender)

0 1
A 0.20104666 0.17226341
D 0.14086350 0.17444396
SA 0.13257741 0.04928042
SD 0.06061928 0.06890536
There are 20% female participants with opinion A. this if followed by 17% males with opinion of A as well as 17% males with opinion of D.
c)
mylogit <- glm(opinion ~ male, data = gsskidvalue, family = "binomial")
summary(mylogit)
Call:
glm(formula = opinion ~ male, family = "binomial", data = gsskidvalue)
Deviance Residuals:
Min 1Q Median 3Q Max
-1.4091 -1.3992 0.9622 0.9707 0.9707
Coefficients:
Estimate Std. E
or z value Pr(>|z|)
(Intercept) 0.50778 0.05895 8.614 <2e-16 ***
male 0.02210 0.08658 0.255 0.799
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 3029.9 on 2292 degrees of freedom
Residual deviance: 3029.9 on 2291 degrees of freedom
AIC: 3033.9
Number of Fisher Scoring iterations: 4
myprobit <- glm(opinion ~ male, family = binomial(link = "probit"), data = gsskidvalue)
summary(myprobit)
Call:
glm(formula = opinion ~ male, family = binomial(link = "probit"),
data = gsskidvalue)
Deviance Residuals:
Min 1Q Median 3Q Max
-1.4091 ...
SOLUTION.PDF

Answer To This Question Is Available To Download

Related Questions & Answers

More Questions »

Submit New Assignment

Copy and Paste Your Assignment Here