Add training workflow, datasets, and runbook

training_data/relevant/text/90f75be472833bacf1d1b7d8efa7caa761fb2260814a4d87139c43dab96e3563.txt

@@ -0,0 +1,37 @@
+Chapter 39: Volatility Trading Techniques 817 
+someone may have inside information that justifies expensive options. This is another 
+reason why selling volatility can be difficult: You may be dealing with far less infor­
+mation than those who are actually making the implied volatility high. 
+COMPARING IMPLIED AND HISTORICAL VOLATILITY 
+The most common way that traders determine which options are cheap or expensive 
+is by comparing the current composite implied volatility with various historical 
+volatility measures. However, just because this is the conventional wisdom does not 
+necessarily mean that this method is the preferred one for determining which options 
+are best for volatility trades. In this author's opinion, it is inferior to the percentile 
+method (comparing implied to past measures of implied), but it does have its merits. 
+The theory behind using this method is that it is a virtual certainty that implied and 
+historical volatility will eventually converge with each other. So, if one establishes 
+volatility trading positions when they are far apart, there is supposedly an advantage 
+there. 
+However, this argument has plenty of holes in it. First of all, there is no guar­
+antee that the two will converge in a timely manner, for example, before the options 
+in the trader's position can become profitable. Historical and implied volatility often 
+remain fairly far apart for weeks at a time. 
+Second, even if the convergence does occur, it doesn't necessarily mean one will 
+make money. As an example, consider the case in which implied volatility is 40% and 
+historical volatility is 60%. That's quite a difference, so you'd want to buy volatility. 
+Furthermore, suppose the two do converge. Does that mean you'll make money? No, 
+it does not. What if they converge and meet at 40%? Or, worse yet, at 30%? You'd 
+most certainly lose in those cases as the stock slowed down while your options lost 
+time value. 
+Another problem with this method is that implied volatility is not necessarily 
+low when it is bought, nor high when it is sold. Consider the example just cited. We 
+merely knew that implied volatility was 40% and that historical volatility was 60%. We 
+had no perspective on whether 40% was high, medium, or low. Thus, it is also nec­
+essary to see what the percentile of implied volatility is. If it turns out that 40% is a 
+relatively high reading for implied volatility, as determined by looking at where 
+implied volatility has been over the past couple of years, then we would probably not 
+want to buy volatility in this situation, even though implied and historical volatility 
+have a large discrepancy between them. 
+Many market-makers and floor traders use this approach. However, they are 
+often looking for an option that is briefly mispriced, figuring that volatilities will