WEBVTT 00:00.870 --> 00:06.300 In this video we will do the same as in the last video simulating one million random portfolios. 00:06.300 --> 00:09.080 But this time we were slightly wary. 00:09.120 --> 00:15.630 The most sensitive fact on our forecast the expected return of our six stocks and actually forecasting 00:15.630 --> 00:22.500 stock returns was a very difficult task and some experts would argue that you cannot forecast them at 00:22.530 --> 00:22.950 all. 00:23.250 --> 00:28.030 Nevertheless there exists a few approaches and methods how to make forecasts. 00:28.080 --> 00:34.390 However this is clearly beyond the scope of this course and in the end it's not an exact science. 00:34.470 --> 00:38.910 So variations of 2 3 or 4 percentage points that can happen easily. 00:39.420 --> 00:41.670 And that's exactly what we'll do in this video. 00:41.970 --> 00:46.890 So we were very aid each and every forecast that returned by 3 percentage points. 00:47.040 --> 00:53.390 And then we will have a look what happens to the efficient frontier and to the max Sharpe Rachel portfolio. 00:53.580 --> 01:01.210 And still we have implied that the summary constituents data frame and on the right hand side we have 01:01.330 --> 01:07.720 the expected returns for our first trial and now it's the second round. 01:08.200 --> 01:13.870 And we very late each expect that returned by three percentage points or families instead of having 01:13.870 --> 01:14.860 25 percent. 01:14.860 --> 01:22.210 We have 22 percent then for Boeing instead of having 15 we have 18 then for this NEI instead of 8 percent 01:22.210 --> 01:29.980 we have 11 and also for IBM we have eleven instead of eight then for Coca-Cola we take 7 percent instead 01:29.980 --> 01:31.100 of 10 percent. 01:31.330 --> 01:33.340 And finally for Microsoft. 01:33.400 --> 01:37.150 We predict that 12 percent instead of 15 percent. 01:37.150 --> 01:38.710 So we created the new column 01:41.470 --> 01:48.760 and one more time we calculate the expected return of one million portfolios by using here the DOT method 01:48.790 --> 01:58.730 on the expected return to her column and we pass again the transpose the weights matrix and then the 01:58.730 --> 02:05.960 next step the calculator the expected risk and also here we use the covariance matrix assuming that 02:05.960 --> 02:11.850 the the variances then to the covariance is off for the past also persist in the future. 02:11.870 --> 02:14.990 So even if we changed here the expected returns. 02:14.990 --> 02:20.450 So the expected risks of our 1 million portfolios do not change. 02:20.600 --> 02:23.210 So this remains unchanged. 02:23.210 --> 02:31.010 And then we also calculate these sharp ratio and we create a summary to data frame with the columns 02:31.010 --> 02:32.950 return risk and Sharpe Ratio 02:36.910 --> 02:42.810 so here we have 1 million portfolios and again we can visualize them. 02:42.820 --> 02:51.840 And this time we create the risk return cloud for our summary to data frame as well as for our first 02:51.840 --> 02:56.300 trial with uh the different return forecasts. 02:56.370 --> 03:05.250 So we have our initial portfolios in red color and now our new portfolios based on our revised the return 03:05.250 --> 03:09.330 forecasts in blue color and let's have a look here 03:15.450 --> 03:20.820 and here we can see the efficient frontier of our first trial and red and the efficient frontier based 03:20.820 --> 03:24.080 on our revised the return expectations in blue. 03:24.480 --> 03:27.510 And secondly a significant difference here. 03:27.930 --> 03:32.880 So we only changed it's an area of return expectations by 3 percentage points. 03:33.450 --> 03:41.250 But this makes actually a huge difference and we can also analyze our revised summary to data frame. 03:41.850 --> 03:46.690 So the max Sharpe ratio has now a sharp ratio of four point eight two. 03:46.710 --> 03:50.730 And we can also search the max Sharpe race for portfolio. 03:50.970 --> 03:56.450 It's with the indexed label eight hundred twelve thousand four hundred fifty two. 03:56.550 --> 04:00.060 And we can also select the max Sharpe Ratio portfolio. 04:00.060 --> 04:05.400 So we have a return of sixteen point eighty three percent and a risk of 18 percent. 04:05.400 --> 04:13.350 And we can compare this to our farmer Max drop rates or portfolio so the difference is actually not 04:13.530 --> 04:14.850 too substantial. 04:14.850 --> 04:20.320 So our farmer Max operational portfolio showed a return of sixteen point one. 04:20.480 --> 04:24.530 And he and our revised the portfolio we have sixteen point eight. 04:24.750 --> 04:28.140 And also the risk is here a slight slightly higher. 04:28.140 --> 04:35.010 But to sum it up to a risk and return of our max operator our portfolio did not change substantially. 04:35.550 --> 04:42.330 However let's have a look at the debates of our new Max Sharpe Ratio portfolio and that's create a panda 04:42.330 --> 04:42.890 series. 04:42.900 --> 04:46.980 So that's the weights of our new Max Sharpe Ratio portfolio. 04:47.340 --> 04:53.720 And let's compare those weights to our farmer Max Sharpe Ratio portfolio and we can see that the weight 04:53.720 --> 04:59.810 of Amazon decrease the slightly from 35 percent to 29 percent. 04:59.810 --> 05:08.720 Then we can see that the weight of the Boeing stock increased substantially from 18 percent to 53 percent 05:09.560 --> 05:17.840 and also the weight of the DNA and the IBM stock significantly increased from actually 0 or 1 percent 05:17.840 --> 05:20.330 to 10 and 7 percent. 05:20.420 --> 05:27.950 And in contrast that the weight off the Coca-Cola stock significantly decreased from 45 percent to 9 05:27.950 --> 05:29.210 percent. 05:29.210 --> 05:35.360 And finally for the Microsoft stock we can see only a little change in the weight but as a summary we 05:35.360 --> 05:43.140 can say that to be only slightly changed our return expectations and also the performance of the max 05:43.140 --> 05:48.680 Sharpe Ratio portfolio in terms of risk and return did not change substantially. 05:48.710 --> 05:54.410 However the rates in our max Sharpe racial portfolio those changed substantially. 05:54.920 --> 05:58.860 And that's another pitfall of using forecast that portfolios. 05:59.210 --> 06:04.910 So the best are the optimal portfolios and the rates and the best and optimal portfolios are highly 06:04.910 --> 06:12.200 sensitive to the return forecasts and this pretty much reduces the practical ability of this approach 06:12.200 --> 06:13.460 here. 06:13.460 --> 06:13.850 All right. 06:13.850 --> 06:20.330 We are finished with this video and then the next video we will summarize the pitfalls of using a forecast 06:20.330 --> 06:22.600 that portfolios for asset allocation. 06:22.940 --> 06:24.770 So hope to see also there by.