Scribd
Upload
14 views2 pages

Econometrics Analysis2forfiverr

Uploaded by

Afra Jagger
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
14 views2 pages

Econometrics Analysis2forfiverr

Uploaded by

Afra Jagger
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
You are on page 1/ 2

1

Submit your writeup with tables and figures in a pdf document, your dataset as a CSV or
XSLX file, your R script or other code in a runnable format or as an addendum to your
writeup.
It should be roughly 5-10 pages including tables and figures.

1. Motivation
a. Start by asking a question about the world, and proposing a hypothesis in the form of
“does X affect Y?” so “an increase (or decrease) in X should lead to an increase (or
decrease) in Y.” (The increase and decrease do not need to match.)

2. Acquiring and accessing data


a. Find a dataset related to the hypothesis you posed in the previous question.
Choose a dataset that interests you. Download and open it in your statistical
software. Explain any transformations or changes you make to your data, if any.
You will need at least 3 variables.
The Federal Reserve Economics Database (FRED) has many economics related datasets.

Yahoo Finance has easily accessible stock data.

Zillow and realtor.com have real estate and home-value data. Data.gov has all sorts of
data
provided by US government agencies. The OECD, UN and IMF have good datasets on
other
countries. Also, many foreign countries have their own statistics organizations, e.g.
Statistics Canada, or the UK national statistics office. The Brooklyn library also has access
to many other useful datasets including Mergent Online for company financial data.

3. Identifying types of data

a. Is the dataset cross sectional, time series, panel, or repeated cross-sectional data?

Give the frequency if your data is measured across time.


4. Data description and summary statistics
a. Find the following statistics for one variable in your dataset. Interpret the mean
and standard deviation of your variable. Include a histogram of your data.
Mean Median
Minimum Maximum
Range Interquartile range
Variance Standard deviation
bonus: use library(“moments”)
Skew Kurtosis
2
5. Hypothesis testing
a. Use one of the tests we discussed in class, or from the textbook, to test the
hypothesis you proposed in the first question. State the test, and null and
alternative hypothesis you intend to test.
6. Simple regression analysis
a. Theoretical model specification
i. Define a simple regression model with predictor and response variables.
b. Goodness of fit / ANOVA
i. Assess your model using: t-tests, f-test, 𝑅2
c. Interpretation
i. Interpret the sign, size, and significance of each parameter in your model.

7. Multiple regression analysis


a. Theoretical model specification
i. Extend the regression model with multiple predictors and justify their
inclusion.
b. Goodness of fit / ANOVA
i. Assess your model using t-tests, f-test, Adj-𝑅2
c. Interpretation
i. Interpret the sign, size, and significance of each parameter in your model.
8. Non-linear regression analysis
a. Theoretical model specification
i. Specify a non-linear regression model and justify its choice.
b. Goodness of fit / ANOVA
i. Assess your model using: t-tests, f-tests, AIC/BIC/HQIC
c. Interpretation
i. Interpret the sign, size, and significance of each parameter in your model.

9. Assessment multicollinearity, heteroskedasticity, and serial correlation.

10. Optional extensions


a. A regression discontinuity assessment
b. Instrumental variables assessment
c. A sigmoid regression analysis
d. ARCH / GARCH / ARIMA time series models

11. Write a description of what you did, and how it shows (or does not show) the relationship
suggested in your hypothesis. Also discuss how this addresses your initial question.
Include a discussion on why you chose the variables you did to measure the effect of your
hypothesis. Consider various biases that may be present in your data.

You might also like