Quantitative and Algorithmic Trading

http://lstm.iupr.com/

Building Fast High-Performance Recognition Systems with Recurrent Neural Networks and LSTM
Thomas Breuel, Volkmar Frinken, Marcus Liwicki

http://www.tradeciety.com/24-statist...rs-lose-money/



“95% of all traders fail” is the most commonly used trading related statistic around the internet. But no research paper exists that proves this number right. Research even suggests that the actual figure is much, much higher. In the following article we’ll show you 24 very surprising statistics economic scientists discovered by analyzing actual broker data and the performance of traders. Some explain very well why most traders lose money.

1. 80% of all day traders quit within the first two years. 1
2. Among all day traders, nearly 40% day trade for only one month. Within three years, only 13% continue to day trade. After five years, only 7% remain. 1
3. Traders sell winners at a 50% higher rate than losers. 60% of sales are winners, while 40% of sales are losers.2
4. The average individual investor under performs a market index by 1.5% per year. Active traders under perform by 6.5% annually. 3
5. Day traders with strong past performance go on to earn strong returns in the future. Though only about 1% of all day traders are able to predictably profit net of fees. 1
6. Traders with up to a 10 years negative track record continue to trade. This suggest that day traders even continue to trade when they receive a negative signal regarding their ability. 1
7. Profitable day traders make up a small proportion of all traders – 1.6% in the average year.However, these day traders are very active – accounting for 12% of all day trading activity. 1
8. Among all traders, profitable traders increase their trading more than unprofitable day traders. 1
9. Poor individuals tend to spend a greater proportion of their income on lottery purchases and their demand for lottery increases with a decline in their income. 4
10. Investors with a large differential between their existing economic conditions and their aspiration levels hold riskier stocks in their portfolios. 4
11. Men trade more than women. And unmarried men trade more than married men. 5
12. Poor, young men, who live in urban areas and belong to specific minority groups invest more in stocks with lottery-type features. 5
13. Within each income group, gamblers under perform non-gamblers. 4
14. Investors tend to sell winning investments while holding on to their losing investments. 6
15. Trading in Taiwan dropped by about 25% when a lottery was introduced in April 2002. 7
16. During periods with unusually large lottery jackpot, individual investor trading declines. 8
17. Investors are more likely to repurchase a stock that they previously sold for a profit than one previously sold for a loss. 9
18. An increase in search frequency [in a specific instrument] predicts higher returns in the following two weeks. 10
19. Individual investors trade more actively when their most recent trades were successful.11
20. Traders don’t learn about trading. “Trading to learn” is no more rational or profitable than playing roulette to learn for the individual investor.1
21. The average day trader loses money by a considerable margin after adjusting for transaction costs.
22. [In Taiwan] the losses of individual investors are about 2% of GDP.
23. Investors overweight stocks in the industry in which they are employed.
24. Traders with a high-IQ tend to hold more mutual funds and larger number of stocks. Therefore, benefit more from diversification effects.

Conclusion: Why Most Traders Lose Money Is Not Surprising Anymore
After going over these 24 statistics it’s very obvious to tell why traders fail. More often than not trading decisions are not based on sound research or tested trading methods, but on emotions, the need for entertainment and the hope to make a million dollars in your underwear. What traders always forget is that trading is a profession and requires skills that need to be developed over years. Therefore, be mindful about your trading decisions and the view you have on trading. Don’t expect to be a millionaire by the end of the year, but keep in mind the possibilities trading online has.
————

– 1Barber, Lee, Odean (2010): Do Day Traders Rationally Learn About Their Ability?
– 2Odean (1998): Volume, volatility, price, and profit when all traders are above average
– 3Barber, & Odean (2000): Trading is hazardous to your wealth: The common stock investment performance of individual investors
– 4 Kumar: Who Gambles In The Stock Market?
– 5 Barber, Odean (2001): Boys will be boys: Gender, overconfidence, and common stock investment
– 6Calvet, L. E., Campbell, J., & Sodini P. (2009). Fight or flight? Portfolio rebalancing by individual investors.
-7Barber, B. M., Lee, Y., Liu, Y., & Odean, T. (2009). Just how much do individual investors lose by trading?
– 8Gao, X., & Lin, T. (2011). Do individual investors trade stocks as gambling? Evidence from repeated natural experiments
– 9Strahilevitz, M., Odean, T., & Barber, B. (2011). Once burned, twice shy: How naïve learning, counterfactuals, and regret affect the repurchase of stocks previously sol.
– 10Da, Z., Engelberg, J., & Gao, P. (2011). In search of attention
– 11De, S., Gondhi, N. R. & Pochiraju, B. (2010). Does sign matter more than size? An investigation into the source of investor overconfidence

Nice article infinitus. Couldn't agree with the research findings more :-)

Hi Guys,

im very interested in quantitative trading and data analysis, I will start to look for it and let you know

Thank you, very nice thread.
You will be taking a lot of my time to dig through this thread. Love this!

Beware of Low Frequency Data


Conclusion

The three example strategies above illustrate that even when a strategy trades at low frequency, maybe as low as once a month, we often still require high frequency TAQ data to backtest it properly, or even economically. If the strategy trades intraday, even if just once a day, then this requirement becomes all the more important due to the flip-flopping of the order book in the millisecond time frame.

Rent, don’t buy, data: our experience with QuantGo (Guest Post)
By Roger Hunter

As a Quant I have to say that you covered this topic very well! Keep it up .

What strategies derived from this approach in forex or equity market?

Nice topic, Interested in such professional approach and wish it will be continued.

My understanding is Quantitative Trading in equity market, you analyse the micro-economy or 'mid scale fundamentals' of business and once you have historical data on return and risk, you setup/adjust the portfolio which is normally a slow pace or long run view. Is this correct that if you derive any strategy from quantitative methods in Forex market, still you need knowledge of fundamentals?

Or is it a matter of mathematical models on random or semi-random data? if so, what data other than price and indicators? any external data?

[quote=infinitus;6768034]

"Trading's not a game – It's an IQ test" {algodude}

What would be my approximately IQ if I have absolute knowledge of movements of prices? Please.

[quote=whateveridon;9230650] The word intelligence comes from the latin intellegere which means to comprehend/understand. This is not the faculty to solve a problem but to understand why the problem is interesting to be solved. If you had absolute knowledge of movements of prices you would be Laplace's demon embodiment. It is supposed to mechanically apply motion rules without a need to understand them. On the other hand, smart or idiot, you would be the wealthiest demon in the world ^^

[quote=PipMeUp;9231281] I in theory I have 100% hits, invest with my theory using "actual" prices is more difficult. I do not believe be Demon of Laplace.

[quote=whateveridon;9231372] A huge company's CFO in Asia shall seal a multi-billion deal with a foreing partner. That's Asian session time and liquidity is poor. He calls his bank to convert big money. His order makes the market move just a few pips lower. This triggers a bunch of SL that pushes the market lower where the liquidity is worse. More stops are hit and trigger sell orders in cascade. People see a breakout on their 5min chart.

A huge company's CFO in Asia shall seal a multi-billion deal with a foreing partner. That's Asian session time and liquidity is poor. He takes a coffee while the price moves a little bit higher. He calls his bank to convert big money. His order makes the market move a bit lower cancelling the last little move up. No SL got triggered. People see nothing on their 5min chart.

Are you telling me you can forecast with 100% acuracy that a CFO is gonna fetch a coffee just by looking at the charts?

I'm not sure I will invest anything in your theory. Sorry.


(NOTE: yes the example is super simplistic but you get the idea...)

[quote=whateveridon;9231372]
at least you can blame failure on quantum mechanics (see Copenhagen interpretation, indeterminacy, coincidence)

sorry for late answer


you can use whatever you like to put in algos - fundamentals, TA, sentiment analysis, ebb and flow, the moon, etc.

http://www.investopedia.com/terms/q/...ve-trading.asp

Excellent thread on the subject of Quantitative Analysis..Thank you, Infinitus

Yes.

I will speak about my Theory little bit.

It is not if you want to believe or not to believe, is about logic. The price of the future that is not yet available in present time (the candles of the prices that will be formed in the future) is going to be predicted with prices of the past, it is a fact.
It is not possible to predict the future with data from the present time, if the current price in is of 11.001 and the price changes to 11.002, it is not possible to predict the trend with only information of the present time that is 11002, if you do not take the previous price 11.001, it is not logical. If you think that I am wrong as a user called Rparm that says that the price of the past has no influence in present and future time, try to predict the tendency just by using the current price.


If I know that the price of the past time influences future time, then I can begin to build a strategy with 100% of hits, I already have a theory.

This is not a quote of mine. That's whateveridon's (http://www.forexfactory.com/showthre...72#post9231372)

I will read your fully stuff in the future.