| StockFetcher Forums · Filter Exchange · Another approach to RSI(2) | << 1 2 3 >>Post Follow-up |
| jldelta 16 posts msg #105696 - Ignore jldelta |
3/27/2012 11:19:46 AM Hello everybody, first-post SF total newbie here, trying to make the move from intraday trading of fast tick charts to EOD trading. I know that several members have generously shared their research into variations of Connors RSI(2). I want to do something slightly different based on Ric Miller’s work at TSM (sorry to mention a commercial service in my first post, but it’s relevant as it will provide my universe of stocks). I am going to use US stocks rather than ETFs & the stock universe will be pre-screened via the TSM fundamental and technical screens on a weekly basis (although many stocks may stay on the Watch List for many weeks at a time). This means manually updating it every Monday morning – I am hoping I can copy>paste the list into SF - but being a Brit currently living in the countryside outside Amsterdam, the European timing works, as does my entry & exit execution to be at the following day’s Open, rather than current day Close. Ric – an apparently rigorous statistician – has determined that this execution on Open barely degrades performance. He has also determined that the premise to use fundamentally strong stocks works even in a terrible year like 2008. So Step 1 for me is to work out how to screen the chosen parameter set, which is : Close above MA(200) RSI(2) below 15 last 2 days, then Buy next day Open RSI(2) above 80, exit half position next day on Open Exit second half position on discretion I don’t think I need to code the exit, as I cannot imagine having a live trade without a chart up & can eyeball it accordingly, so the code snippet I need is how to code the two consecutive days’ closes below 15. Is it really as simple as : Close above MA(200) and RSI(2) has been below 15 for the last 2 days ? I regret I haven't worked out how to correctly wrap code on this forum yet. Then for the chart : And draw MA(200) And draw RSI(2) And chart length is 6 months (I am a former TradeStation “cut-and-paste” non-coder only) I know this may seem ridiculously simple, but the instruments will have already been through multiple techno-fundamental screens and I am just looking for this specific pullback. I intend to forward, rather than backtest this from 2 April (a true backtest would have to be one week at a time...), & also research the use of options as opposed to shares. I’m familiar with using deep ITM calls as a stock surrogate, but not for what are typically very short-term trades of a few days, so whereas in the past I have tended to never hold an option less than 30 days out, it may be that shorter-term options are more appropriate because of potentially very short time decay & low premium. I wonder if any members care to share their experience of this ? If options do turn out to be preferable, then I need to add a line of code that filters any non-optionable instruments (p85 of the manual, I see...). Anyway, it’s very KISS, the essential idea being the use of fundamentally pre-screened stocks. J. |
| Kevin_in_GA 4,599 posts msg #105697 - Ignore Kevin_in_GA |
3/27/2012 11:58:01 AM So Step 1 for me is to work out how to screen the chosen parameter set, which is : Close above MA(200) RSI(2) below 15 last 2 days, then Buy next day Open RSI(2) above 80, exit half position next day on Open Exit second half position on discretion I don’t think I need to code the exit, as I cannot imagine having a live trade without a chart up & can eyeball it accordingly, so the code snippet I need is how to code the two consecutive days’ closes below 15. Is it really as simple as : Close above MA(200) and RSI(2) has been below 15 for the last 2 days ? I regret I haven't worked out how to correctly wrap code on this forum yet. Then for the chart : And draw MA(200) And draw RSI(2) And chart length is 6 months +++++++++++++++++++ You almost wrote the code here already. Now if you are going to use a custom stock list, the best way to do that is to create a watchlist for the prescreened stocks, and dump the list into it. Than add to the top of your filter the following line - "apply to watchlist(XXX)" where XXX is the name of the list you created. That way, the filter will only look at the universe of stocks you selected from Ric's method. Kevin |
| jldelta 16 posts msg #105698 - Ignore jldelta |
3/27/2012 12:50:14 PM Very helpful indeed, thank-you, Kevin. |
| jldelta 16 posts msg #105781 - Ignore jldelta modified |
4/2/2012 10:32:33 AM Day 1 222 TSM Weekly picks this morning, 19 of which show up on my filter screen. Too many... moving to Plan B : "What Ric did" here (back test of 90 TSM regular Weekly picks over both 2008 and 2009) : http://www.tradingmarkets.com/stocks/trading-lessons/buying-weakness-and-selling-strength-684961.html Edit : how do I post wrapped code WITHOUT generating charts that do not relate to what I'm doing (as they are not on my Watchlists), please ? |
| jldelta 16 posts msg #105782 - Ignore jldelta modified |
4/2/2012 10:38:39 AM Plan B produces 4 selections : CDNS +0.42% TEX +2.58% AKRX +1.20% HPY +3.33% Total +7.53% Edited for Day 1 returns. I am not intending to maintain this thread on a regular basis : next Step is to work out an efficient, effective & VIABLE way to trade the signals. |
| kobe0129 1 posts msg #105798 - Ignore kobe0129 |
4/3/2012 3:48:30 PM What stocks are on your watch list? |
| Kevin_in_GA 4,599 posts msg #105799 - Ignore Kevin_in_GA |
4/3/2012 8:54:12 PM Plan B produces 4 selections : CDNS +0.42% TEX +2.58% AKRX +1.20% HPY +3.33% Total +7.53% +++++++ Uh, not quite ... you'd want the AVERAGE of those four trades (+1.88%) rather than the sum. |
| jldelta 16 posts msg #105807 - Ignore jldelta |
4/4/2012 6:00:15 PM Oops... thank-you, Kevin. @kobe0129 - the list changes every weekend; as it's from the subscription-based service mentioned in the first post, I won't list the current one. You could do somethng similar using one of the IBD screens, for example, although the TSM screening process includes some IBD content. Using options was a great idea, but I'm realising a lot of the stocks don't have the large option volume that's needed for efficient trading of small moves, so it looks like trading the shares may be the only sensible way to go. I'm now runnng both versions of the Filter, ie. <5 and <15 having realised that there is an additional, discretionary filter to apply : the pullback must be to an obvious area of support, so a larger number of screened picks may be preferable. |
| wkloss 231 posts msg #105835 - Ignore wkloss |
4/5/2012 6:16:32 PM jldelta Could you share any information you have about this service with regard to results? I looked at the website and my 1st impression was that they provide you a list of candidates and trade instructions. If that is correct, what will you do with the SF filter? Bill |
| saico 59 posts msg #105896 - Ignore saico |
4/12/2012 3:06:20 PM I'm trying to figure out, if it would have a negative performance impact in the long run to close positions on indraday rsi readings above 70, instead of waiting for the rsi to close above 70 at the end of a day. Any input is much appreciated. Saico |
| StockFetcher Forums · Filter Exchange · Another approach to RSI(2) | << 1 2 3 >>Post Follow-up |
Copyright 2022 - Vestyl Software L.L.C.•Terms of Service | License | Questions or comments? Contact Us
EOD Data sources: DDFPlus & CSI Data
Quotes delayed during active market hours. Delay times are at least 15 mins for NASDAQ, 20 mins for NYSE and Amex. Delayed intraday data provided by DDFPlus