Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit

Assignment 1 (4070)

Note:

1. Total 40 marks.

2. You also need to submit the STATA do file and log file. Please combine all files into a single PDF file and submit it through Blackboard.

1. (20 marks)

Suppose that you are interested in studying the factors determining wages. You plan to estimate a wage equation using December 2021 Labour Force Survey (LFS) data. You would like to use ln(wage) as the dependent variable. For explanatory variables, you consider the following variables: Age, Gender, Education, and Experience.

a. Find the corresponding data in the LFS for the variables listed below. For example, if you find a variable named HRLYEARN (usual hourly wage) in the LFS, and you decide that you will use HRLYEARN for wage in you regression model, you should fill the first row as follows: (5 marks)

The name of the variable in LFS

Label of the variable in the LFS

Wage

HRLYEARN

usual hourly wage

Age

Gender

Education

Experience

b. List at least three specifications that you can think of for estimation. For example, the simplest specification could be:

List the estimation results for all the specifications in a table. For example, the table could be like the one below: (5 marks)

Dependent variable: ln(wage)

Variables

Coefficient Estimates and (Standard Errors)

Specification A

Specification B

Specification C

Specification D

C

1.35

(0.27)

Age

0.038*

(0.041)

Gender

0.0045**

(0.0017)

Education

0.05***

(0.008)

Experience

-0.00066*

(0.00009)

SSE

237.98

0.2696

*** Significant at the 1 percent level

** Significant at the 5 percent level

*Significant at the 10 percent level

Note the estimated values listed for Specification A are hypothetical, not the real estimates based on LFS.

c. Based on the results, summarize your findings. For example, you can discuss things such as whether experience matters for determining wages; how education affects wage rate; whether there is a gender wage gap based on your estimation; and other related issues. (10 marks)

2. (20 marks)

Suppose that you are interested in finding out the relationships between aggregate consumption and output, and between aggregate investment and output for the Canadian economy for the sample period of 2000Q1 to 2021Q3.

a. Using Data (formerly CANSIM), find data for aggregate output, consumption, and investment. Note that you need to use real, seasonally adjusted series for these three variables. Fill in the table below. For example, the first row shows that for output, you use Gross domestic product at market prices, and it is from Table 36-10-0104-01 in Data (380-0064 in CANSIM). (5 marks)

Variables in Casim

Data Table

output

Gross domestic product at market prices

36-10-0104-01

Consumption

Investment

b. Use Table: 17-10-0009-01 (formerly CANSIM 051-0005) to find data for the Canadian population series for the sample period, compute per capita output, consumption, and investment. Fill out the table below. (5 marks)

Mean

Per capita output

Per capita consumption

Per capita investment

c. Using nonstationary time series in regression can lead to spurious regression. Detrending the data (converting nonstationary series to stationary series) before running regression is a common practice. In what follows, you are asked to conduct the following two basic detrending methods: linear detrending and first differencing.

Take the natural logarithm of the per capita output, consumption, and investment, i.e., compute y=ln (per capita output), c=ln (per capital consumption), and i=ln (per capita investment). You are asked to conduct the following detrending exercises on y, c, and i.

1) Linear detrending: assume that there is a linear trend for each of the data series. Remove the linear trend from the data. For the detrended data, compute the standard deviation of each time series. Which series has the highest standard deviation? Compute the correlations between output and consumption, and between output and investment. Is consumption procyclical or not? What about investment? (5 marks)

2) Another way of having a stationary series is to take the first difference of the original data. That is, you need to generate a lagged variable for each of the series and then take the first difference. For example, for output, you need to generate. Once you have generated the detrended data after first differencing, repeat the exercises regarding standard deviation and correlation in 1). (5 marks)