Linear Regression Standard Analyses

Statistics and Econometrics

Lab Exercise:

Linear Regression

 Standard Analyses

To hand in

You don’t to submit this lab!

Learning objectives

The main objective of this computer lab is designed to provide students with the basic tools to work with data using the open source package Gretl.

After completing the lab the student shall be able to:

1. Be familiar with the Workspace in Gretl;
2. work with adding new variable, computing log, lag
3. work with graphs
4. Be familiar with hypothesis tests, Transformation, Normality tests
5. Formulate a simple linear regression model and explain how the model  parameters can be interpreted;

Amir:   a.sadoghi@fs.de

Data:

You can either use dataset or download data and build your own dataset

The file CAPM_with_FFF.csv contains daily stock prices of market mover companies (Ford, Microsoft, Facebook, Bank of America Corporation , Citigroup Inc) and index values for the S&P500 and factors from Fama French Factors: (SMB stands for "Small [market capitalization] Minus Big" and HML for "High [book-to-market ratio] Minus Low")  from 1/6/2012 to 28/02/2014.

Download the daily stock prices of market mover companies (Ford, Microsoft, Facebook, Bank of America Corporation , Citigroup Inc) and index values for the S&P500  from .[ http://finance.yahoo.com]  and three-month T- bills from June 1, 2012 to today  from https://research.stlouisfed.org/fred2/series/DTB3/downloaddata   

Assignment 1:

 Descriptive Statistics

1. Obtain summary statistics of daily price of Ford, Microsoft, Facebook and index values for the S&P500 (e.g. Mean, Median, Minimum, Maximum, Standard deviation, Skewness, kurtosis, percentiles) Write a summary of your findings based on this data.
2. Plot time series of daily price Ford, Microsoft, Facebook and index values for the S&P500and explain your finding.
3. Plot histogram and boxplot of daily price of Ford, Microsoft, Facebook.
4. Compute correlation between daily price of Ford, Microsoft, Facebook and index values for the S&P500
5. Compute first lag of daily price of index values for the S&P500, compute correlation between daily index values for the S&P500 and its first lag.

Returns 

1. Compute the log returns for stocks and S&P500 and plot time series of log returns and explain your finding. (Clean data for missing values)

In Gretl, from main menu choose “Add”, “Define new variables” Generate a new variable; use this box to define a new variable, on the pattern name = formula.