Matlab Assignmen - Mhat and Sdhat
Essay by kirill.talay • January 30, 2016 • Case Study • 753 Words (4 Pages) • 956 Views
MatLab Assignment 2
Edward Mouradian
Question 1.1
Part A
- The momentum check script starts by loading the prices & assets data.
Then it calculates the simple returns for the prices. It selects the
individual subsets of columns, moves to selecting the portfolio columns,
choosing the most extreme assets to form portfolios. After that it creates
Weighted matrices. It calculates the portfolio returns and finally
it runs the FNsumstatsNEW script to produce the statics for each
market index (DJINDUS & FRCAC40) also giving us the high and low MOM.
[pic 1]
T represents the periods in the sample data in this case a time span of 10 years.
Mhat and sdhat represent the estimator of the mean and the standard deviation respectively in the data set and in this case they are positive for assets in question. Gaussian distribution has a kurtosis of 3 and a skewness of 0. Both DJINDUS and FRCAC40 the skewness is close to 0 which means that it is symmetrical. While kurtosis of DJINDUS and FRCAC40 are really high which means the data has fat tails and low even distribution.
JB represents a test statistics of Jarque-Bera test which states that residuals are normally distributed. The low P-value (less than 5%) means that H0 does not hold, thus residuals of the regression are not normally distributed.
The P-values overall hold the hypothesis.
T represents the periods in the sample data in this case a time span of 10 years.
Mhat and sdhat represent the estimator of the mean and the standard deviation respectively in the data set and in this case they are positive for assets in question. Gaussian distribution has a kurtosis of 3 and a skewness of 0. Both DJINDUS and FRCAC40 the skewness is close to 0 which means that it is symmetrical. While kurtosis of DJINDUS and FRCAC40 are really high which means the data has fat tails and low even distribution.
JB represents a test statistics of Jarque-Bera test which states that residuals are normally distributed. The low P-value (less than 5%) means that H0 does not hold, thus residuals of the regression are not normally distributed.
The P-values overall hold the hypothesis.
T represents the periods in the sample data in this case a time span of 10 years.
Mhat and sdhat represent the estimator of the mean and the standard deviation respectively in the data set and in this case they are positive for assets in question. Gaussian distribution has a kurtosis of 3 and a skewness of 0. Both DJINDUS and FRCAC40 the skewness is close to 0 which means that it is symmetrical. While kurtosis of DJINDUS and FRCAC40 are really high which means the data has fat tails and low even distribution.
...
...