StockFetcher Forums · View by Author: (850 messages)  [ Display By: Date / Subject ]<< 1 ... 81 82 83 84 85 >> 
Filter Exchange · Breakout Trading Filter - GREAT RESULTS!
nibor100
msg #132948
12/1/2016 5:46:16 AM

Hey Wilson,

I mistakenly was looking at the number from the 80 day close line instead of the 80 MA line in my last post on chart CISG.

So how about looking at the ELOS chart at day 2/1/16, mine shows 200 ema at 8.66, 200 ma at 9.22, for avg of 8.94, with 80 ma at 7.48 putting the avg about 19.5% above the 80 day ma.

I'm relatively new to Stockfetcher so thanks for the chart mov avg Settings tip, should make my future filters shorter.

Sorry about the earlier confusion ,
Ed S.



Filter Exchange · Breakout Trading Filter - GREAT RESULTS!
nibor100
msg #132931
11/30/2016 2:50:04 PM

Wilson,

I'm fairly certain now that is not the case due to looking at a one year graph of CISG after adding the following 3 lines to your filter

draw ema(200)
draw ma(200)
draw ma(400)

Looking at the 1 year chart in SF 2.0, so I can see the values for every line plotted on the chart for any point by hovering, it appears that on March 17, 2016 the 80 MA was at 10.17 and the avg of the ema200 and ma200 was at 8.14 which is a diff of much more than 15%.

I believe this is because the coding in your filter for v2a is identical to that for v3a, so the lookback period for both is the same duration and starts on the same date as is true for v3a and v3b.

I was going to backtest this in another program I have access to and I want to make sure of all of the parameters of your filter that supported your paper trade results that you posted earlier so I'll have something to compare to.

Thanks,
Ed S.

Filter Exchange · Breakout Trading Filter - GREAT RESULTS!
nibor100
msg #132911
11/30/2016 11:42:05 AM

Wilson,

In comparing your breakout filter to your description paragraph it appears there might be typos for variables v3a and v3b as they seem to duplicate v2a and v2b making it possible for the totals to hit 240 but it appears the count statements are only covering 160 days of history in total.

Could you verify?

Thanks,
Ed S.

Filter Exchange · A NEW ^VIX TRADING SYSTEM
nibor100
msg #132782
11/25/2016 11:33:42 AM

Using the past 6 months of ^VIX data downloaded from Yahoo Finance into Excel, I determined that if one went short each day, then 85.8% of the time there would be a profitable exit possible at an Open, within the next 10 days.

If one went long each day then 82.5% of the time there would be profitable exit possible at an Open, over the following 10 days.

However, if one went both long and short each day then just 68% of the time would both trades have a possible profitable exit within the next 10 days.

Possibly shows how regularly volatile the ^VIX is when Open prices are used for entries and exits.
Ed S.



Filter Exchange · A NEW ^VIX TRADING SYSTEM
nibor100
msg #132739
11/23/2016 1:28:08 PM

Kevin,

Regarding the last 8 trade results you recently posted, I can only verify that in following a 5 unit portfolio approach that 2 of the larger gaining trades would not have been executed, due to date overlap, similar to early October when there was also 7 short signals close together.

Ed S.

Filter Exchange · A NEW ^VIX TRADING SYSTEM
nibor100
msg #132737
11/23/2016 1:17:21 PM

dtatu,

Could you post the dates of entry and exit that were used by you for the 3 test methods whose 13 results you posted earlier so we can determine the corresponding Kevin signals, etc.?

You stated the "system" works after using 3 methods trading ~4 trades each and I notice that your $ profits related to %profits are spread quite a bit as a 7.16% gain got $1,070.80, and a 0.68% gain got $686.86.

Is your version of the "system" alternating between those 3 methods or have you settled on one method going forward?

Thanks,
Ed S.



Filter Exchange · A NEW ^VIX TRADING SYSTEM
nibor100
msg #132734
11/23/2016 1:07:08 PM

Miketranz,

In Kevin's first post, the 2nd backtest on SS covering a different span of 10 years, the results were a bit further away from ~90% winners down to ~71% winners.

Ed S.

Filter Exchange · A NEW ^VIX TRADING SYSTEM
nibor100
msg #132572
11/16/2016 1:49:42 PM

Some observations from my SS trial and SF filter coding learning experience over the past 2 weeks:

a. Based on just the last 2 months of Long trades results from SS the avg days held for winners has been 4.5 days which is very close to Kevin's SS backtest avg for winning trades of 4 days held. I was curious about the days held because the losing trades avg days held for his backtest was a much higher 7 days

The 8 long trades since the system has gone live, have had 7 winners and 1 small loser for a win rate of 88%..

b. The original backtest was reported to have a 10 days in trade limit, but the SS results show that trades of 11 days in duration occur. Probably because the exit code has "($daysheld > 10)" instead of '>9'.

c. There have been a few reported signal anomalies throughout this thread regarding SS, and my tests seem to confirm that there was no long entry for Long#1 on .9/6 nor for Long#9 on 9/13 (though both of these false signals with the trade exit dates reported ended up profitable). There was also an unreported Long#1 entry for 10/20 that exited on 10/28 for an 8.59% gain.

d. In my early filter testing I was stumped by the ^VIX chart not showiing up in my results which SF Support verified by stating "StockFetcher does not include indices in results due to the lack of volume availability.".
I've since discovered a workaround, if I click on any stock chart returned as a filter result and I enter ^VIX in the chart name box on upper left and click chart the ^VIX chart will be displayed.(this holds true for SF 2.0 also)

e. Yesterday, I asked SF Support what Limit value they use for the ASI, since Long#1 SS code uses an ASI(100), and my testing seemed to indicate SF uses a Limit value of 30, but some of the results were inconsistent; and I also asked them how far back in the past does SF go in order to start the cumulative count as starting date can drastically effect cums.

Their response "The actual starting point for cumulative measures is typically at least 100 days (usually more). The starting point will also differ for charts vs. filter results."

That last part was very surprising to me and seems a bit odd, has anyone run across that before where chart values differ from filter values? .

Thanks,
Ed S.






Filter Exchange · A NEW ^VIX TRADING SYSTEM
nibor100
msg #132375
11/6/2016 1:10:00 PM

Kevin,

I decided to try coding the price channel screen portion of VIX LONG SIGNAL #1 in SF to see if I could get it to work properly and to better learn the SF coding language I got it to work after several tries so that it matched what Stock Charts explains as the methodology for Price Channels.

Yesterday I started my SS 30 day trial so I could check on how SS calculates Price Channels, after following your instructions for downloading ^VIX price data and creating a single symbol sector and I was able to verify by using the SS charts that SS calculates Price Channels in the same manner as Stock Charts. com explains and also that the results matched my SF filter results So I was feeling fairly confident.

Then I pasted your SS code snippet ; ' (pcu(15)-pcl(15) < mov(pcu(15)-pcl(15), 15, simple))' into the Indicator Entry box for adding a Chart Value in SS and it also turned out to match that portion of my SF filter.

However, neither my SF filter nor the SS chart show a buy signal for the Price Channel portion of VIX LONG SIGNAL #1 for Sep 2nd which is one of the long signals you reported in your msg of Sep 5th.
Since VIX LONG SIGNAL #1 has 3 conditions separated by 'AND' I'm assuming the Price Channel portion must signal a Buy along with the other 2 signals or its not a signal.

Is there an easy way you can verify that signal again for Sep 2nd and provide the values for pcu(15),pcl(15),(pcu(15)-pcl(15), and mov(pcu(15)-pcl(15), 15, simple) so I can double check my results.

That would be a big help,
Ed S.


Filter Exchange · A NEW ^VIX TRADING SYSTEM
nibor100
msg #132276
11/1/2016 11:39:49 AM

Unless the full backtest has been re-run for portfolios sizes of 1 and 2, I suspect greater returns are not guaranteed; due to the much greater # of trades that would be missed as a result of such small portfolio sizes. The odds of missing some of the larger gaining trades would probably increase.

Ed S.

StockFetcher Forums ·  · << 1 ... 81 82 83 84 85 >>

*** Disclaimer *** StockFetcher.com does not endorse or suggest any of the securities which are returned in any of the searches or filters. They are provided purely for informational and research purposes. StockFetcher.com does not recommend particular securities. StockFetcher.com, Vestyl Software, L.L.C. and involved content providers shall not be liable for any errors or delays in the content, or for any actions taken based on the content.


Copyright 2021 - 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


This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.