Simulation and VaR

In the previous sections, we have learned that there are two ways to estimate VaR for an individual stock or for a portfolio. The first method depends on the assumption that stock returns follow a normal distribution. The second one uses the sorted historical returns. What is the link between those two methods? Actually, Monte Carlo simulation could be served as a link. First, let's look at the first method based on the normality assumption. We have 500 Walmart shares on the last day of 2016. What is the VaR tomorrow if the confidence level is 99%?

#
position=n_shares*x.close[0] 
mean=np.mean(ret)
std=np.std(ret)
#
VaR=position*(mean+z*std)
print("Holding=",position, "VaR=", round(VaR,4), "tomorrow")
('Holding=', 26503.070499999998, 'VaR=', -641.2911, 'tomorrow')

The VaR is $641.29 for tomorrow with a confidence level of 99%. Here is how Monte Carlo simulation works. First, we calculate the mean and standard...