Add training workflow, datasets, and runbook

training_data/relevant/text/e4675f125ec3edf6031cf18a2117fe6a83388f9a4cd1097a1fda894d5b44ab82.txt

@@ -0,0 +1,36 @@
+466 Part IV: Additional Considerations 
+taking a moving average of the last 20 or 30 days' implied volatilities. An alternative 
+that does not require the saving of many previous days' worth of data is to use a 
+momentum calculation on the implied volatility. For example, today's final volatility 
+might be computed by adding 5% of today's implied volatility to 95% of yesterday's 
+final volatility. This method requires saving only one previous piece of data - yester­
+day's final volatility - and still preserves a "smoothing" effect. 
+Once this implied volatility has been computed, it can then be used in the 
+Black-Scholes model ( or any other model) as the volatility variable. Thus one could 
+compute the theoretical value of each option according to the Black-Scholes formu­
+la, utilizing the implied volatility for the stock. Since the implied volatility for the 
+stock will most likely be somewhat different from the implied volatility of this par­
+ticular option, there will be a discrepancy between the option's actual closing price 
+and the theoretical price as computed by the model. This differential represents the 
+amount by which the option is theoretically overpriced or underpriced, compared to 
+other options on that same stock. 
+EXPECTED RETURN 
+Certain investors will enter positions only when the historical percentages are on 
+their side. When one enters into a transaction, he normally has a belief as to the pos­
+sibility of making a profit. For example, when he buys stock he may think that there 
+is a "good chance" that there will be a rally or that earnings will increase. The investor 
+may consciously or unconsciously evaluate the probabilities, but invariably, an invest­
+ment is made based on a positive expectation of profit. Since options have fixed 
+terms, they lend themselves to a more rigorous computation of expected profit than 
+the aforementioned intuitive appraisal. This more rigorous approach consists of com­
+puting the expected return. The expected retum is nothing more than the retum that 
+the position should yield over a large number of cases. 
+A simple example may help to explain the concept. The crucial variable in com­
+puting expected return is to outline what the chances are of the stock being at a cer­
+tain price at some future time. 
+Example: XYZ is selling at 33, and an investor is interested in determining where 
+XYZ will be in 6 months. Assume that there is a 20% chance of XYZ being below 30 
+in 6 months, and that there is a 40% chance that XYZ will be above 35 in 6 months. 
+Finally, assume that XYZ has an equal 10% chance of being at 31, 32, 33, or 34 in 6 
+months. All other prices are ignored for simplification. Table 28-5 summarizes these 
+assumptions.