tag:blogger.com,1999:blog-19505137.post389916503091406994..comments2016-08-14T03:34:09.589-05:00Comments on TraderFeed: Determining Price Targets for TradesBrett Steenbarger, Ph.D.http://www.blogger.com/profile/11988667917563876202noreply@blogger.comBlogger6125tag:blogger.com,1999:blog-19505137.post-43190800526604084582014-04-22T01:11:38.690-05:002014-04-22T01:11:38.690-05:00your blog is interesting but your posts are too lo...your blog is interesting but your posts are too long doc. I am a party time trader and it is part of my routine to read a couple of market related blogs everyday before I leave for work and i would like to add your blog to this repertoire but your posts are far from being straight to the point. You often spend a couple of paragraphes explaininbg something which could be explained in just a few words. Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19505137.post-12033647883503653032014-04-21T03:10:20.600-05:002014-04-21T03:10:20.600-05:00Row 8 of Atr Target Test Indicator has look-ahead ...Row 8 of Atr Target Test Indicator has look-ahead bias - you can't using today's values of indicators in calculations until end of today.<br />Hence, something like it would be correct:<br />Atrs = AvgTrueRange(AtrLength)[1];yuprikhttps://www.blogger.com/profile/17148676660065843172noreply@blogger.comtag:blogger.com,1999:blog-19505137.post-20520769307806438532014-04-18T22:09:09.758-05:002014-04-18T22:09:09.758-05:00I'm not sure why a day before a holiday would ...I'm not sure why a day before a holiday would indicate that the market would be unlikely to trend to a far target, as often before holidays that's not an uncharacteristic market behavior.<br /><br />Again, I'd suggest that by tracking one-touch binaries that one could see the real probabilities of hitting a level in real-time. I have considered to do this as an adjunct to aide me in hitting my targets. The NADEX binaries give real-time odds on the market closing at or above a certain level. And, what may be relevant and make the approach described here less useful then one might anticipate is that the expected value will increase with the commensurate decrease in the probability. <br /><br />What would be useful would be a spreadsheet or a mathematical formula (Easylanguage) for estimating one-touch probabilities without needing the manual input of volatility levels. It should be possible to produce target, stop probabilities from the one-touch estimation. I believe binomial tree is one method of estimation, and I'm sure there are others.<br /><br />Dr. Brett, as well, you may find it interesting that my understanding of volume based on my deep subjective and equally sophisticated objective studies doesn't necessarily agree with your findings & beliefs. I do, however, find the "real-time" estimate of hitting targets intriquing. Curtishttps://www.blogger.com/profile/01381976399305765020noreply@blogger.comtag:blogger.com,1999:blog-19505137.post-85660050462548127002014-04-18T16:58:06.417-05:002014-04-18T16:58:06.417-05:00Once you add a stop loss to your trades, these pro...Once you add a stop loss to your trades, these probabilities are unlikely to be representative of the actual probability of your meeting your target. <br /><br />I'm not sure how useful it might be but you could take a look at NADEX binary probabilities which show the real-time probability of CLOSING at or above a certain level. <br /><br />These aren't the "one touch" probability though-- which is what you'd want for a futures target stop loss and target. <br /><br />Curtishttps://www.blogger.com/profile/01381976399305765020noreply@blogger.comtag:blogger.com,1999:blog-19505137.post-47747671130826667512014-04-18T11:25:01.843-05:002014-04-18T11:25:01.843-05:00Thanks, David; I appreciate your putting the work ...Thanks, David; I appreciate your putting the work into that. I will redo calculations and repost shortly!<br /><br />Brett Brett Steenbarger, Ph.D.https://www.blogger.com/profile/11988667917563876202noreply@blogger.comtag:blogger.com,1999:blog-19505137.post-41210329272649020402014-04-18T10:41:23.730-05:002014-04-18T10:41:23.730-05:00The current 5 day ATR on SPY is 2.03 and your 78% ...The current 5 day ATR on SPY is 2.03 and your 78% point is 0.74 (187.13-186.39), or 0.3645 times the Atr. The 78% probability sounded low to me so I did a test with the TradeStation indicator code below.<br /><br />It counts the total number of bars and the number of bars (hits) where today's high or low meets or exceeds yesterday's close by +/- the PctgOfAtr Atrs. It then calulates the percent of hits.<br /><br />I ran the test over the entire SPY history (back to 1994). Since the late 1990s the percent of the time that +/- 0.3645 of the 5 day Atr threshold is at least met is over 94% with the latest being 94.77%. In fact +/- 0.5 Atrs is hit 81.5% of the time.<br /><br /><br />--------------------------------------------------------------<br /><br />Atr Target Test Indicator<br /><br />1. Create new indicator and add code below<br /><br />2. Set properties:<br /> Hit and Miss are histograms, PctHit is a line<br /> Hit is Green, Miss is Red, PctHit is Yellow<br /><br />3. Verify code (compiles it)<br /><br />4. Insert into chart<br /><br /><br />code begins below this line:<br /><br />inputs:<br />AtrLength(5),<br />PctgOfAtr(.3645);<br /><br />variables:<br />TotBars(0),<br />TotHits(0),<br />Atrs(0);<br /><br /><br />Atrs = AvgTrueRange(AtrLength);<br /><br />if H >= C[1] + Atrs * PctgOfAtr or L <= C[1] - Atrs * PctgOfAtr then<br /> begin<br /> TotHits = TotHits + 1;<br /> Plot1(100, "Hit");<br /> end<br />else<br /> Plot2(-100, "Miss");<br /><br /><br />TotBars = TotBars + 1;<br />Plot3(100 * TotHits/TotBars, "PctHit");<br /><br />Anonymoushttps://www.blogger.com/profile/13011439402730498022noreply@blogger.com