Kevin_in_GA
4,599 posts
msg #137401
- Ignore Kevin_in_GA |
8/6/2017 7:54:59 PM
I download SOXS every night - I did see a price error back in 2012 so I make the first date 1/2/2013. Try that and see if it helps. No issues with recent data.
|
ferndave
65 posts
msg #137410
- Ignore ferndave |
8/7/2017 9:27:45 AM
Kevin: I have been using the first date of 1/1/2014. I just tried clearing prices, symbols, and sectors on a different StrataSearch iteration and it is still missing 7/31 data for a good chunk of ETFs. Even deleted a few from Price and Symbol Setups with the same results after a fetch. At a loss as to why it isn't working.
|
mahkoh
1,065 posts
msg #137421
- Ignore mahkoh |
8/7/2017 1:06:18 PM
Your importfile does not have rows with only a date, does it? Because that would overwrite the data with empty cells.
|
ferndave
65 posts
msg #137423
- Ignore ferndave |
8/7/2017 2:33:21 PM
mahkoh: no lines with only dates.
I think I figured out a fix. I tried a fresh Google import, so it was missing the 7/31 data. I made a csv with the missing 7/31 numbers for the symbols I need. Looks like this:
Symbol,Date,Open,High,Low,Close,Volume
SOXL,2017-07-31,95.5,96.790001,92.25,93.169998,432400
etc.
Did a separate ascii import and it filled in the missing values from the file. I think my problem was using the filename option rather than getting the symbol from within the file. I can schedule this to run right after the main import. Hopefully this will get me back to more of an automated set-up.
Thanks for the help and suggestions.
|
pthomas215
1,251 posts
msg #137598
- Ignore pthomas215 |
8/12/2017 6:28:07 PM
so, could someone please articulate what functionality stratasearch has that improves trading judgement. I havent used it but think i will get a cheap laptop to toss it on.
|
sandjco
648 posts
msg #137613
- Ignore sandjco
modified |
8/14/2017 10:17:16 AM
Hey PT...IMHO...
1. SS needs to run on a good PC (RAM intensive). A cheap PC may not cut it if you want to use it while SS is running in the background.
2. SS allows you to build your own system; it can back test your variables and/or parameter variables
3. It has a scanner
As a newbie, I was able to build my first crude system and back test in less than 10min by following the video tutorials. However, the "lingo" to code isn't as simple as the ones in SF (for someone like me). The actual backtesting of it, however, took days (I have 8GB DDR3 running on 8 Cores at 3.5Ghz as a perspective).
I'm sure there are a lot more features in the program that I have yet to discover. It does require a lot of reading the manual as the tabs or buttons aren't as intuitive as the ones in SF. For example, when you want to create your own "watchlist", SS calls it "sectors" I think.
Impressive piece of software at a price you just cannot say no to! Thank you SS and thanks Kevin for bringing this up.
Hope that helps.
|
pthomas215
1,251 posts
msg #137614
- Ignore pthomas215 |
8/14/2017 10:20:56 AM
sandjco, it helps a ton. thank you.
|
shillllihs
6,044 posts
msg #137615
- Ignore shillllihs |
8/14/2017 10:38:59 AM
Exactly why I don't expend that much energy. It is true, the spoils go to those who put in the effort, but for me, it's when I put in too much effort and it fails and it just makes you want to walk away from everything permanently. My whole life everything has just come ez to me. So I look for the easiest route. It's just me.
|
dgkaufman
23 posts
msg #137626
- Ignore dgkaufman |
8/14/2017 1:09:19 PM
I actually tried coding the Shills $1 filter in SS over the weekend. Still a work in progress.
I have been trying to code the majority of the filters I have found such as the 5% weekly filter but have not had too much success translating the language yet.
I know Kevin has posted several of the coding in the forums but if anyone wants to collaborate on coding some of these filters, I am more than happy to help.
|
ferndave
65 posts
msg #137663
- Ignore ferndave |
8/15/2017 3:15:31 PM
@pthomas215: SS is like allowing all of the technical indicators found on SF to be brute force run against a specified date range, and list of stocks, or just one. The result is a list of filters that meet the criteria you set. For example, the resulting list of filters, using different combinations of indicators and parameters, could all be profitable 90% of the time, with trades averaging less than 15 days, and drawdowns limited to -15%.
Every night you run those filters against updated daily data. It will tell you if you should buy/sell/hold at the next open.
Kevin is easily the most experienced and can speak to its effectiveness over time.
|