Friends, thanks for your reading and comments, DW team really appreciates your time. Therefore we decided not to review all modern portfolio theories and prove math models incompetence because other teams had already written about it. We just only Konstantin Tzarikhin's works followers and we totally agree that it's impossible to create true comparable investment portfolio model with precise forecasts in case of risks variety and black swans. By the way, people like figures and they try to describe environmental process by math, but nowadays global financial market is over the math. Far too often figures are used for proving previously accepted decision that is why many algorithms, maybe apart from arbitrage strategies, are fundamentally wrong. I do no how to finish this passage. So, ladies and gentlemen, let us present DW investment portfolio. How we chose stocks We decided to take automotive industry because our friends are keen on cars and we didn't believe in the first-round passing without its' support (11 and 12 places). Then we started to select shares by the regional and operating criteria. According to these criteria we chose to include into the watch-list: Toyota (TM), Fiat (FCAU), Thor (THO), Daimler (DAID), Nissan (NISANF), AVTOVAZ (AVAZ), Ford (F), Volkswagen (VOW) and Hyundai (HYMTF). After the first round we found that several automotive stocks are fundamentally sick and coupled with contest rules we had to choose companies from the other industries which would potentially suit for us. That is why the key criteria for the second round stocks were: NYSE listing, strong profitability ratios, 1-month long-trend in the stock's industry, low β-ratio. As a result we included: Ennis (EBF), Gilead (GILD), JPMorgan (JPM), 3M (MMM) and Surgutneftegas (SNGS). How we chose portfolio Finally we had 15 shares which trades on 3 different exchanges. In case of USA strength, we took NYSE as a mother-boerse, and all trades from the MOEX and DAX we exchanged according to the Forex. Then we downloaded average historical prices from the 01.01.15 - 04.12.15 and exported them into EXCEL. Table 1. DW 15 Stocks Historical Prices (01.01.15 - 04.12.15) After this step we started to choose portfolio model. So, we got:duration: 3 monthsprofitability: more, than S&P 500 ( we chose this index as a market profitability) and STOXX® Europe 600 Automobiles & Partsothers: low risks, low β As it was previously mentioned, we came up with idea that there is no ideal model and the majority of them are based on Markowitz Theory. That is why excluded all synthetic calculations. We took average prices (OHLC/4) of the day because this price includes daily volatility of the share. Table 2. DW Profitability, Standard Deviation and β We added historical prices of S&P 500 and STOXX® Europe 600 Automobiles & Parts and calculated an average profitability and according to this ratio we excluded Ford, 3M, Volkswagen and Hyundai. Later we estimated standard deviation σ, which estimates the movement of the stock.We can't use σ as risk ratio at whole because the risk definition depends on the the market deal (long/short), therefore this risk figure could surplus or diminish our results like multiplier. The final ratio is β, which estimates the market dependency of the stock. For instance, the key idea of the β-portfolio is to combine independent from the market yield stocks. Here you can see, that we didn't filter stocks by standard deviation and β is in suitable limit. So, this is the final stock-list of the portfolio:TM - Toyota (auto)FCAU - FCI (auto)THO - Thor (auto)GM - General Motors (auto)EBF - Ennis (consumer goods)GILD - Gilead (healthcare)JPM - JPMorgan (finance)DAID - Daimler (auto)NSANF - Nissan (auto)AVAZ - AVTOVAZ (auto)SNGS - Surgutneftegas (oil&gas) Table 3. DW Complete Correlation Matrix This matrix shows, Russian stokes (AVAZ and SNGS) are the most independent stocks, whereas USA carmakers (THO and GM) and JPMorgan Chase & Co have the closest connection. That is why the best way to create portfolio is to include Russian stocks and companies with relatively low correlation. Table 4. DW Covariation Matrix This matrix was used for estimating the share of every stock in investment portfolio and potential risks for the next 3 months. Table 5. DW Investment Portfolio Model In this case we tried to forecast a low-risk portfolio. To sum up, we calculated profitability and roughly risks per every share. Covariation and correlation matrix determined an optimal structure with low risks and comparable yield. How we estimated our portfolio Nothing special, there are two indicators groups:Risk RatiosProfitability Ratios Table 6. DW Risk Ratios Standard deviation is an ordinary multiplier, which in our super negative scenario we use it as a portfolio risk. It estimates 1,05% which is 0,08% more than S&P 500. There is nothing significant in this figure which is standard for the market. Sharp Ratio - is a measure for calculating risk-adjusted return, and this ratio has become the industry standard for such calculations. It estimates 2,98% which shows that portfolio management is quite effective especially in comparison with Automotive industry. β Ratio - is a measure of the volatility, or systematic risk, of a security or a portfolio in comparison to the market as a whole. It estimates 0,85% which is lower the average market level which is good for our strategy. Treynor Ratio measures returns earned in excess of that which could have been earned on a riskless investment per each unit of market risk. We have 0,037 which underlines that our portfolio is 0,037 more profitable, than the whole market. This is a pay for low risks. Jenson Alpha Ratio - A risk-adjusted performance measure that represents the average return on a portfolio over and above that predicted by the capital asset pricing model (CAPM), given the portfolio's beta and the average market return. This is a backward ratio which shows that portfolio yield is 0,59% less the average market results. Table 7. DW Yield Ratios The chart below shows yield of every stocks in portfolio. Ennis is the most yield estimated 8,86%, Fiat (4,2%) and Surgut (4,2%) are 3rd and 2nd respectively (Chart 1). For instance, S&P 500 yield is 0,42%. Chart 1. Assets Portfolio Yield The chart below shows, that total portfolio yield estimates 3,54% which is 2,84% and 3,12% more, than S&P 500 and Zero-Risk Asset respectively (3-month US Treasuries). Chart 2. 3-month Portfolio, S&P 500 and Zero-Risk Asset Yield Chart 3. 3-month Investment Portfolio In conclusion, I would like to add that our forecast isn't precise because we used only Markowitz and beta models which were created in the last century. In fact, we have a country diversified stocks from 5 sectors with low market dependency and correlation to each other. At the last moment we decided not to include Automotive Index as a result of industry diversification. This is the end of our pilot portfolio, if you have several questions we are happy to answer them.