The project task is: “Find stocks that give solid and stable profits in the long run.” To accomplish this task we need to make decisions. I describe how the project approaches the task and by which means the task is solved.
If you have read the introduction to this project and the description of the project mission, you might remember the words “decision making”, “stock rating”, “financial key metrics” and “stock prices”, as well as “scoring methods” and “data analytics”.
In this section, the meaning of these words is explained and related to the accomplishment of the task which is: “Find stocks that give solid and stable profits in the long run.”
First of all, in this context, “decision making” means choosing stocks worth buying to accomplish the task. There are several stages of decision making during the overall process. We want to make proper decisions to choose the right stocks at all these stages. “Stock rating” is the tool we apply to improve our decision making along the way.
The first step of decision making relates to the stock rating strategy we apply to the whole set of stocks. We decide how the strategy is designed to extract a subset of stocks from the total set. There might be stocks which we want to exclude from the beginning on, so we filter them out. According to the concrete strategy design different remaining stock subsets are obtained. This is what we call “prefiltering”.
The second step of decision making relates to the “stock rating” rules which are applied to the remaining subset of stocks from step one. These rules use “financial key metrics” and “stock prices” to calculate a number, the “total score”. The total score is calculated by classifying stock feature values into three classes, the outperformer, the neutral and the underperformer class, each class having a numerical value of +1, 0 and -1 respectively and then summing up the (weighted) results. This classification is the essential part of “scoring methods” which are part of the field of “data analytics”. Depending on how the stock rating rules are designed, a different total score is obtained. The higher the total score the higher the chance of success should be.
Various stock rating strategies can be designed. Designs are different when they comprise different stock features or rate the same features higher or lower. According to the chosen features and how they are rated, a stock rating strategy might be more or less successful in accomplishing the task.
The third step of decision making relates to the choosing of stocks from the rated subset of all stocks. The intuitive and straightforward approach is to choose a certain number of stocks with the highest score, e.g. the first 20 to 30 stocks. Another approach is to further examine the stocks, e.g. according to single feature values or in comparison to industry sector competitors. Also it might be reasonable to factor in future growth estimates of the companies or their industry sector.
To close this section, the approach of this project is to apply stock rating strategies for better decision making and task accomplishment. The prefiltering and calculation of total scores is done automatically using software. The obtained total scores are published on this site. The performance of different stock rating strategies is evaluated and monitored to check how well the stock rating strategies are able to accomplish the task.