Add training workflow, datasets, and runbook

training_data/curated/text/9cf639880bc226199732ec50e773d32a98bd053fc1331e2b305672ac14116f74.txt

@@ -0,0 +1,29 @@
+Finding Mispriced Options    • 153
+“We think that these prices far below the current price are much more 
+likely than they would be assuming normal percentage returns. ” (Or, in a 
+phrase, “We’re scared!”)
+If we compare the delta-derived “cone” with a theoretically derived 
+BSM cone, here is what we would see:
+Oracle (ORCL)
+Date
+Price per Share
+60
+50
+40
+30
+20
+10
+-
+6/21/201612/24/20156/27/201512/29/20147/2/20141/3/20147/7/20131/8/20137/12/2012
+Of course, we did not need the BSM cone to tell us that the points 
+associated with the downside strikes look too low. But it is interesting to see 
+that the upside and most likely values are fairly close to what the BSM projects. 
+Note also that the downside point on the farthest expiration is nearly 
+fairly priced according to the BSM, contrary to the shorter-tenor options. 
+This effect could be because no one is trading the far ITM call long-term 
+equity anticipation securities (LEAPS), so the market maker has simply 
+posted his or her bid and ask prices using the BSM as a base. In the market, 
+this is what usually happens—participants start out with a mechanically 
+generated price (i.e., using the BSM or some other computational option 
+pricing model) and make adjustments based on what feels right, what 
+arbitrage opportunities are available, and so on.