9792024鍾安民

1) 程式碼(歷史模擬法)

series VaR=NA

smpl full

smpl 1 1900

loop for j=1901..1974

    scalar qt=quantile(Y,0.01)

    #---k2 is last smpl

    scalar k2=$t2+1

    #---calculate relative value at risk

    VaR[k2]=-qt

smpl +1 +1

end loop

smpl full

結果：1901-1974 數值為 1.460950

series VaR=NA

smpl full

smpl 1 1900

loop for j=1901..1974

scalar w0=1000000

#---sigma is standard error

    scalar sigma=sd(Y)

    #---normal distribution with 0.01% critical value

    scalar c_005=critical(z,0.05)

#---k2 is last smpl

    scalar k2=$t2+1

#---calculate relative value at risk

    VaR[k2]=w0*c_005*sigma/100

smpl +1 +1

end loop

smpl full

執行結果 VaR(mean)