Add training workflow, datasets, and runbook

training_data/curated/text/5ccff9a01f18f72cad18a47ba29c38e3aff195a52dc886c8412ebad299d5890f.txt

@@ -0,0 +1,39 @@
+Chapter 27: Arbitrage 427 
+generally declared a special dividend in the fourth quarter of each year, but that its 
+normal quarterly rate is $1.00 per share. Suppose the special dividend in the fourth 
+quarter has ranged from an extra $1.00 to $3.00 over the past five years. If the arbi­
+trageur were willing to speculate on the size of the upcoming dividend, he might be 
+able to make a nice profit. Even if he overestimates the size of the special dividend, 
+he has a limited loss. Suppose XYZ is trading at 55 about two weeks before the com­
+pany is going to announce the dividend for the fourth quarter. There is no guarantee 
+that there will, in fact, be a special dividend, but assume that XYZ is having a rela­
+tively good year profitwise, and that some special dividend seems forthcoming. 
+Furthermore, suppose the January 60 put is trading at 7½. This put has 2½ points of 
+time value premium. If the arbitrageur buys XYZ at 55 and also buys the January 60 
+put at 7½, he is setting up a risk arbitrage. He will profit regardless of how far the 
+stock falls or how much time value premium the put loses, if the special dividend is 
+larger than $1.50. A special dividend of $1.50 plus the regular dividend of $1.00 
+would add up to $2.50, or 2½ points, thus covering his risk in the position. Note that 
+$1.50 is in the low end of the $1.00 to $3.00 recent historical range for the special 
+dividends, so the arbitrageur might be tempted to speculate a little by establishing 
+this dividend risk arbitrage. Even if the company unexpectedly decided to declare no 
+special dividend at all, it would most likely still pay out the $1.00 regular dividend. 
+Thus, the most that the arbitrageur would lose would be 1 ½ points (his 2½-point ini­
+tial time value premium cost, less the 1-point dividend). In actual practice, the stock 
+would probably not change in price by a great deal over the next two weeks (it is a 
+high-yield stock), and therefore the January 60 put would probably have some time 
+value premium left in it after the stock goes ex-dividend. Thus, the practical risk is 
+even less than 1 ½ points. 
+While these types of dividend risk arbitrage are not frequently available, the 
+arbitrageur who is willing to do some homework and also take some risk may find that 
+he is able to put on a position with a small risk and a profitability quite a bit larger 
+than the normal discount dividend arbitrage. 
+There is really not a direct form of dividend arbitrage involving call options. If 
+a relatively high-yield stock is about to go ex-dividend, holders of the calls will 
+attempt to sell. They do so because the stock will drop in price, thereby generally 
+forcing the call to drop in price as well, because of the dividend. However, the hold­
+er of a call does not receive cash dividends and therefore is not willing to hold the 
+call if the stock is going to drop by a relatively large amount (perhaps ¾ point or 
+more). The effect of these call holders attempting to sell their calls may often pro­
+duce a discount option, and therefore a basic call arbitrage may be possible. The arbi­
+trageur should be careful, however, if he is attempting to arbitrage a stock that is