Search results

Van Tharp & Ken Long. The SQN "formula" can be found in a few on Van's books. The formula suffers from using the number of trades, IMO.

Winrate is really meaningless by itself. The first "single metric" I would suggest to look at is the profit factor P/F. But single metrics are quite deceptive, especially for large backtests, as they completely hide the variability of performance. Looking at the distribution of a particular...

IMO the number of trades is not enough in itself ... there should also be a condition on the time period those trades cover. I find 10+ years a pretty good benchmark (where applicable), and I would now completely discard anything below 5 years. For me, those figures apply for the backtesting...

The real questions though (wrt fitting), are: 1. how complex is that one-line formula. 2. how complex are the rules to buy or sell around that one-line.

The same question goes for anything really ... whether football or backet-ball or hockey players, objects like boats / cars / bikes / pens / computers / phones, books & software, food, art pieces, and of course companies. It really takes an open market to "objectively" put a price tag on...

It is a volume-based timeframe, for which each bar represents 100 contracts traded.

If you are referring to this system's use of a 100-volume timeframe, I would say the timeframe resolution has nothing to do with the "level" of the signals generated. I use 100-volume to get adequate precision for measuring both volume & time when creating the H/L pivots used to "decode"...

I have already minimized my dependencies to Ninja by creating an abstraction layer between Ninja & my own stuff (that became obviously necessary when I started creating reusable trading components, which where meant to be used from both "indicators" & "strategies"). So let's say I am...

I have no idea what those errors can be. Most often I check my attachments after posting, never found any issue. Does this attachment give you an error? Funny how a P&L curve can tell a story better than the corresponding performance report. P/F 1.38 isn't exactly great, neither is 51%...

I posted in another thread that, to me, backtesting has value only if it shares the same code-path as live execution. If you are not going automated, then IMO backtesting is a waste of time ... - because you won't be able to trade live all the setups that would be found later in backtesting...

Re. plateform ... AmiBroker, NinjaTrader, MultiChart, TradeLink, etc. ... they all have pros & cons. none of them is perfect. You need to know what's important for you, in the short-term & in the long-run, and analyze those plateforms in this light. I can say for myself, object-oriented...

Take-two ... What I am most worried about is missing a "big" thing (the blank item 5) ...), but so far so good. I am on Ninja, using Kinetick(aka IQfeed) as primady datafeed, with accounts at IB & Dorman. 1) 24/7 operations For the last 6 months, for my current systems I have been using...

Sorry folks, I was in the middle of a very long & detailed response, and just lost it all (my bad - I am tired). Thank you for all the suggestions & comments, I will get back to this after a good night of sleep - I need it :(

I have been working on a prototype system which is best described as "always-in". In very simple terms, it uses only 2 kinds of signals, "Long" and "Short", both to close/reverse the current position as well as possibly scale-in the current position. A new position is always started with 1...

Yes, markets are constantly changing, however they are not going to email you next time they change - as a result, the best is to ensure your strategy can survive or better adapt to varying market conditions, and for this you need a lot of historical data. I am an automated trader, so I...

You appear to be just starting ... please consider at least using proper tools if you ever want to have a chance at succeeding. Requirement #1: lots of historical data for the markets you are considering backtesting - 10+ years is ideal - how many 1min bars does this make for just one symbol...

Misaki, you're too funny ... you gave plain bad advice on CPU & RAM, and won't admit to it. Anyone who says #core is less important than CPU frequency for strategy development & backtesting/optimization in Ninja has severe issues. The attached picture shows Ninja running 94% on 12 cores...

This is mostly BS from the uninformed. RAM and CPU are very important for Ninja, if you are doing optimizations on intraday data (enough of it) you *will* need 64GB RAM. Ninja uses all cores in optimization, I personally have a i7-3930k w/ 64GB RAM and use about 50GB RAM for any optimization (CL...

Recurring periods of time during which a strategy produces results "far away" from its nominal results.

From the description of it, IMHO there isn't much hope. You would need to be colocated at the exchange, analyze the order book at ultra-fast speed, to get a chance at this game.