What is this? Risks and? danger
Risk: the risk of misjudgment, that is, the first type of error, also known as the first type of error. When the process involved is still under control, but a certain point falls outside the control limit due to accidental reasons, it is wrong to draw the conclusion that the process is out of control. Such mistakes will lead to unnecessary search for reasons for non-existent problems and increase costs.
Risk: that is, the risk of misjudgment, that is, the second type of error, also known as the second type of error. When the process involved is out of control, but the point caused by accidental reasons still falls within the control limit, it is concluded that the process is still under control. The loss at this time is caused by the increase of unqualified products that have not been detected. The risk of the second kind of error is a function of the following three factors: the interval width of control boundary, the degree of process out of control and the size of subgroup. The nature of the above three factors determines that the risk of the second type of error can only be estimated generally.
Alpha strategy involves a wide range of market fields, and has applications in various markets such as stock market, bond market and commodity market. At present, the most common hedging strategy in the domestic market is alpha hedging strategy, which usually uses the advantages of stock selection and timing to find a spot portfolio with stable excess returns, and separates beta through derivatives such as stock index futures, thus obtaining alpha returns with low market correlation. Especially in a bear market or consolidation period, can it be used? Spot bulls+futures bears? On the one hand, the long position of the portfolio that can obtain excess returns is established, on the other hand, the short position of the stock index futures is established to hedge the systemic risk of the spot portfolio, so as to obtain positive absolute returns. The best test of alpha strategy is the stock selection model.
When constructing individual stock portfolio, investors can use quantitative model to select stocks. Multi-factor model is the most widely used quantitative model for fund companies. Combining these factors organically constitutes a score of each stock, which is also called the expected return of the stock. When determining the weight of individual stocks in the portfolio, professional institutions will select appropriate factors from the quantitative factor library according to the investor's income requirements and risk tolerance, tailor the expected return model for investors, and take into account the expected risk and transaction cost of stocks, and calculate the weight of each stock in the portfolio through optimization methods.
Alpha strategy should essentially belong to a classification of statistical arbitrage. Through the multi-factor regression analysis of historical data, the targets that will perform better than similar stocks in the market are screened out, and the stock index futures are used to hedge and isolate the influence of all stocks rising and falling together on the position value, so as to achieve the purpose of obtaining stable absolute returns. In the China market, many Alpha strategic fund managers privately admit that the market value factor is far more important than any other factor. If we build a portfolio with equal rights, buy all non-ST stocks in the two cities, and sell the Shanghai and Shenzhen 300 Index with the same market value on June 5438+ 10/day, because the market value of these more than 2,000 stocks is much smaller than that of the heavyweights, this portfolio is actually buying a large number of small and medium-sized stocks and selling heavyweights. The absolute return of such a portfolio in the past two years has almost outperformed all Alpha strategic private placements.
Therefore, the view is that almost all Alpha models give the result that the stocks that are most likely to give excess returns in the two cities are concentrated in various emerging industries such as media, computers and medicine, while the industries with large-cap stocks such as financial real estate resources can hardly find any expectation of excess returns. As a result, two problems are exposed: first, some fund managers maintain a large risk exposure. How long and how short are the stocks with non-market value? Packing? Absolute returns unaffected by market fluctuations? . Another problem is that the stocks selected by the same model are too homogeneous, and trampling often happens in extreme cases.