convert stock cusips to stock symbols
Another way is to directly get the list of cusips from SEC published 13f lists here (in pdf) :
13f lists
Quantitative Futures, stocks and Options Trading



Plot of SampleSize Required vs Probability:
T-distribution table VS Sample Size and closeness to Standard Normal:
Survey Example: Some tests for Sample Sizes:
Reverse Experiment: calculating Percentages Needed with a given SampleSize to have statistical significance.
Plot of SampleSize Required vs Probability:
figure()
plot(SampleSizeRequired(:,1),SampleSizeRequired(:,2))
xlabel('Probability P')
ylabel('Sample Size Required')
% T n
[4.303 3
2.228 11
2.086 21
2.042 31
2.00 61];
% As can be seen above Starting from about 21 we can see it get closer to 2
Assume IID (Independent Identical distribution). So now let us say we
need to do a survey. Let us assume that we have
ONLY 21 (From the above Table) people to survey from.
IT IS TIE (50-50). I.e Even though 60% of the people say A, we want to make sure it is different from being a TIE (50-50)
when a bigger population is considered.
% Now lets put some values into the equation:
P = 0.60; p0 = 0.50;%( NULL Hypothesis)
n = 21;
t_stat = (P - p0)/sqrt(p0 * (1-p0)/n)
t_stat = 0.9165
% We GOT 0.9165 Which is LESS than 2. So we FAIL to reject the NULL
% Hypothesis. So we cannot say with certainty that it is NOT a TIE.
% Now lets increase the n to 30.
n=30;
t_stat = (P - p0)/sqrt(p0 * (1-p0)/n)
t_stat = 1.0954
% We GOT 1.0954 Which is STILL LESS than 2. So we FAIL to reject the NULL
% Hypothesis. So we cannot say with certainty that it is NOT a TIE.
% Now lets go further and calculate it for a bunch of n
n = 30:10:100;
t_stat = (P - p0)./sqrt(p0 * (1-p0)./n);
figure()
plot(n,t_stat)
xlabel('Sample Size')
ylabel('T stat')
% BINGO: We got it. So at a sample Size of 100, we get Exactly what we
% wanted. We REJECT the NULL Hypothesis that It was a TIE and declare A as
% our winner in the survey.

% So here we need to find out for what P value is needed.
% With Some Linear Algebraic Manipulations, we get:
% Now find out for different n , what P is needed..
n = 21:30;
P = p0 + [2 * sqrt(p0 * (1-p0)./n)];
figure()
plot(n,P)
xlabel('Sample Size')
ylabel('Probability Needed to have statistical siginificance')
% For Larger Samples:
n = 30:1000;
P = p0 + [2 * sqrt(p0 * (1-p0)./n)];
figure()
plot(n,P)
xlabel('Sample Size')ylabel('Probability Needed to have statistical siginificance')
n = 25 has a truly statistical justification. At n = 25 the Law of Large numbers will start to show a pronounced symmetric/normal distribution of the sample means around the population mean. This normal distribution becomes more pronounced as n is increased.
n = 30 comes from a quote from Student (Gosset) in a 1908 paper "On the probable error of a Correlation" in Biometrika. In this paper he reviews the error associated with drawing of two independent samples from infinitely large population and their correlation (not the individual errors of each sample relative to the sample mean and the population mean!). The text reviews
different corrections to the correlation coefficient given various forms of the joint distribution. In a few sentences, Student says that at n = 30 (which is his own experience) the correction factors don't make a big difference.
Later, Fisher showed that the sample for a correlation needs to be determined based on a z-transformation of the correlation. So, Student's argument is only interesting historically. Also, Student wrote his introduction of the t-test in Biometrika during the same year (his
prior article). Historically, the n = 30 discussed in his correlation paper has been confused with the t-test paper, which only introduced the t-statistic up to sample size 10.
3. http://www.itl.nist.gov/div898/handbook/ppc/section3/ppc333.htm
4. http://www.itl.nist.gov/div898/handbook/prc/section2/prc222.htm
5. http://www.itl.nist.gov/div898/handbook/prc/section2/prc243.htm
This code implements the Black and Litterman Model As given in the paper He & Litterman: The intuition Behind Black- Litterman
Model Portfolios.
% First The Inputs:
% Correlation Matrix: Page 18
Corrmat = ...
[1,0.4880,0.4780,0.5150,0.4390,0.5120,0.4910;
0.4880,1,0.6640,0.6550,0.3100,0.6080,0.7790;
0.4780,0.6640,1,0.8610,0.3550,0.7830,0.6680;
0.5150,0.6550,0.8610,1,0.3540,0.7770,0.6530;
0.4390,0.3100,0.3550,0.3540,1,0.4050,0.3060;
0.5120,0.6080,0.7830,0.7770,0.4050,1,0.6520;
0.4910,0.7790,0.6680,0.6530,0.3060,0.6520,1;];
% Risk Aversion Parameter---Page 10
RiskAversion = 2.5;
% Standard Deviations and Market Capitalization Weights--Table 2 Page 19
stdevs = ...
[16.0000
20.3000
24.8000
27.1000
21.0000
20.0000
18.7000]./100;
MktWeight = ...
[ 1.6000
2.2000
5.2000
5.5000
11.6000
12.4000
61.5000]./100;
tau = 0.05;
Equation 2 Page 3 PI = RiskAversion * Covariance * MktWeight
% Need to Convert Correlation into Covariance
Covmat = Corrmat .* (stdevs * stdevs');
% Now we have everything to implement the Equation
%EqRiskPrem = RiskAversion * Covmat * MktWeight;
EqRiskPrem = RiskAversion * Covmat * MktWeight;
% Print Out Table 2 in Page 19
AssetNames = {'Australia','Canada','France','Germany','Japan',...
'UK','USA'};
Table2 = [{'Assets' 'Std Dev' 'Weq' 'PI'};
{'------' '------' '---' '--'};
AssetNames' num2cell([stdevs MktWeight EqRiskPrem]*100)]
Table2 =
'Assets' 'Std Dev' 'Weq' 'PI'
'------' '------' '---' '--'
'Australia' [ 16] [ 1.6000] [3.9376]
'Canada' [20.3000] [ 2.2000] [6.9152]
'France' [24.8000] [ 5.2000] [8.3581]
'Germany' [27.1000] [ 5.5000] [9.0272]
'Japan' [ 21] [11.6000] [4.3028]
'UK' [ 20] [12.4000] [6.7677]
'USA' [18.7000] [61.5000] [7.5600]
%View1 is The German Equity Market Will Outperform the rest of European
% Markets by 5% a year.
P = [ 0 %Australia
0 %Canada
-29.5 %France
100 %Germany
0 %Japan
-70.5 %UK
0]'./100; %USA
Q = 5/100;
% The Black Litterman Expected Returns are Calculated
% By Equation 8 and the Optimal Portfolio Weights are
% Calculated By
%RiskAversion * Covariance * ExpectedReturns(MU)
% Lambda = 0.302; Page 11
Omega = diag(diag(P*tau*Covmat*P'));
% Equation 8--Expected Returns : MU
PostCov = inv(inv(tau*Covmat) + (P' * inv(Omega) * P));
SigmaP = Covmat + PostCov;
ExpRet=inv(inv(tau*SigmaP)+P'*inv(Omega)*P)* ...
(inv(tau*SigmaP)*EqRiskPrem +P'*inv(Omega)*Q);
% ExpRet = inv(inv(tau*Covmat) + P' * inv(Omega) * P) * ...
%(inv(tau*Covmat) * EqRiskPrem + P' * inv(Omega) * Q);
% Optimal Weights
OptimalWeights = (1/RiskAversion)* inv(SigmaP) * ExpRet;
Tab4Col4 = OptimalWeights - (MktWeight)/(1+tau);
Table4 = [{'Assets' 'P' 'MU' 'W' 'W - Weq/1+tau'};
{'------' '--' '---' '--' '-------------'};
AssetNames' num2cell([P' ExpRet OptimalWeights ...
round(Tab4Col4 * 1000)./1000]*100)]
Table4 =
'Assets' 'P' 'MU' 'W' 'W - Weq/1+tau'
'------' '--' '---' '--' '-------------'
'Australia' [ 0] [ 4.3328] [ 1.5238] [ 0]
'Canada' [ 0] [ 7.5838] [ 2.0952] [ 0]
'France' [-29.5000] [ 9.2991] [-4.0555] [ -9]
'Germany' [ 100] [11.0615] [35.7733] [ 30.5000]
'Japan' [ 0] [ 4.5087] [11.0476] [ 0]
'UK' [-70.5000] [ 6.9550] [-9.7178] [ -21.5000]
'USA' [ 0] [ 8.0756] [58.5714] [ 0]
Published with MATLAB® 7.6