Sensex: Quantifying Sell in May

There is a very famous investment philosophy to Sell in May and return in October.

Lets see how that strategy works for our Indian Markets.

The conditions are to Sell on first day in May and Buy on first day on October with commissions of 5 bps.
We are testing this from year 1980 and 2010 so that it covers most of the history available.

How are the results? Well, its not profitable. The initial equity of $10K got reduced to $9.8K while buy and hold was massively up by ~2 times.

Here's the snap shot of the performance.

There were only 10 profitable years among the 31 years of testing with the year 2010 contributed massively $4780  to the results.
The profits are completely skewed by a single year.

This clearly shows that the strategy is not profitable in itself but there was one profitable trend. That was the higher probability of a dip in early May which lasts for 2-3 weeks on an average.

Here's the equity graph.

Quanting India: Sensex Mean Reversion study

Markets display a very peculiar phenomenon called Mean reversion. Without this there the prices would be just random. See more on this concept at Wikipedia. Link here.

There can be multiple variations of this concept which are applied to markets. Like closing price relation to high or low etc.

One widely followed is the mean reversion to moving averages like 50, 100 and 200 day MA. There are already well defined trading strategies based on the price and the moving averages.

For today let us consider the 50 Simple Moving Average and Sensex. I am taking Sensex because it has longest price history in Indian indices so the strategies can be back tested well.

To look for mean reversion to 50 DMA lets plot the Sensex closing w.r.t 50 DMA.

We get the following chart:

There are few spikes but more or less the prices have stayed within a band. Doing the statistical analysis to the values we get the following:

The values are consistent with the bullish bias of Indian markets for last 30 years.
Going further I calculated the lows of the data set in the bull phase and high points in the bear phase. These values can tell us the probable points where the buying in done in bull and selling in bear phase. And this particular trade has the highest probability of success till our view of market phase is correct.

Well now how to trade with the above strategy at this point.
Looking at the extreme right hand side of the chart of last 2 years we can have some idea about the current setup. The ratio is at the lows of the bull run where generally the buying comes.


Does this indicate that buying can be done at this point. Well there is still no confirmation in the reversion strategy as of now as there are a few elements missing for triggering buy at these levels.

I will update in future posts if there is any indication in that respect.

The above post is meant for quant strategy discussion rather than any investment advice.

Year 2010 Nifty: System testing the Color code

For long I have been writing about the color of the indices which defines the trend.
I think the best way to know about any strategy is to backtest the strategy. So let us test the Nifty for the color system.

Time frame: Daily for year 2010. Testing only for Longs no Short trade tested.

Buy Condition: Buy color
Sell Condition: Neutral or Sell color

Commissions: 0.10% of trade.


The results

The system was equal to buy and hold but the profit factor of the trade at 4.7 makes it attractive with only 10 long trades for the year.

Also the maximum draw down for a single trade was about 3% which is reasonable given that we are trading on daily timeframe.

More details later.

Bank Nifty: System Testing Trend Defining Line

Last month I wrote about a simple concept of buying and selling based on a buying and selling on cross of a simple trend line. Post here.

Then I tested the Nifty daily data of last 15 years and the system was outperforming the buy and hold. Post here.

Today I am testing the same concept for Bank Nifty on weekly charts. The main reason I am testing on weekly instead of daily charts is that the Bank Nifty being a high beta index has more whipsaws on daily charts and the test conditions being very simple it generates a lot of trades some even by just 2 points lower close.

Also there are always different system performing well on different timeframes and we should always trade them in that very timeframe for maximum advantage.

Please see this post as an endeavor to quantifying some technical strategies and then optimize them for better trading results.

Lets look at the results:

Buy & Hold out performance: 54%
Profit Factor (Profit/ Loss):      5.3
Absolute Drawdown:              513 points

Time frame: Daily for last 10 years on Bank Nifty spot.

Buy Condition: Close above the TDL.
Sell Condition: Close below TDL

Short Condition: Close below TDL
Profit Book Condition:  Close above the TDL.

Commissions: 0.10% of trade.
Trade Entry: Buy/Sell on next open bar i.e. if signal is generated before today's close then buy/sell tomorrow open. This is to have a realistic results and avoids preempting the signal though we loose a lot by waiting for next candle.

As can be seen these are very simple rules to trade.

Output

Equity curve is attached.

* CNX BANK INDEX (70001968) Simulation Date 20/01/2011 12:17:04 PM 577 Weekly Bars 07/01/2000 Through 19/01/2011 (4030 Days) Points Only Test

Performance Profit 14724.5876 Pts Performance N/A Annualized Performance N/A Buy & Hold Profit 9559.2855 Pts Buy & Hold Performance N/A Buy & Hold Annualized Performance N/A Trade Summary Total Trades 31 Trade Efficiency -3.59 % Average Profit/Average Loss N/A Profitable Trades Total 15 Long 10 Short 5 Average Profit 1206.4348 Pts Highest Profit 3907.3321 Pts Lowest Profit 32.2313 Pts Most Consecutive 2 Unprofitable Trades Total 16 Long 5 Short 11 Average Loss -210.7459 Pts Highest Loss -830.6802 Pts Lowest Loss -25.8716 Pts Most Consecutive 5 Maximum Position Excursions Long Favorable 5257.9317 Pts Short Favorable 4073.1338 Pts Long Adverse -931.1746 Pts Short Adverse -958.5298 Pts Trade Efficiency Average Entry 60.11 % Average Exit 36.30 % Average Total -3.59 % Average Long Entry 71.38 % Average Long Exit 42.19 % Average Long Total 13.57 % Average Short Entry 49.54 % Average Short Exit 30.78 % Average Short Total -19.69 %

Performance Indices Buy & Hold Index 54.03 % Profit/Loss Index 81.37 % Reward/Risk Index 96.63 % Accounting Initial Equity 0.0000 Pts Trade Profit 18096.5213 Pts Trade Loss -3371.9338 Pts Commissions 115.0702 Pts Interest Credited 0.0000 Pts Interest Charged 0.0000 Pts Final Equity 14724.5876 Pts Open Positions 0.0000 Pts Account Variation Highest Account Balance 15565.2663 Pts Lowest Account Balance -513.0161 Pts Highest Portfolio Value 5257.9317 Pts Highest Open Drawdown -513.0161 Pts Highest Closed Drawdown -461.0274 Pts Account Events Margin Calls 0 Overdrafts 0 Profitable Timing Average Trade Length 30 Longest Trade Length 76 Shortest Trade Length 5 Total Trade Length 451 Unprofitable Timing Average Trade Length 6 Longest Trade Length 21 Shortest Trade Length 1 Total Trade Length 104 Out of Market Timing Average 7 Longest 21 Total 22

System Testing Trend Defining Line

I did a simple system trade on the trend defining line (TDL from now onwards) as described here.

The idea was to buy and sell on cross up or down.
Brief summary:

Time frame: Daily for last 15 years on Nifty spot.

Buy Condition: Close above the TDL.
Sell Condition: Close below TDL

Short Condition: Close below TDL
Profit Book Condition:  Close above the TDL.

Commissions: 0.10% of trade.
Trade Entry: Buy/Sell on next open bar i.e. if signal is generated before todays close then buy/sell tommorrow open. This is to have a realistic results and avoids preempting the signal.

As can be seen these are very simple rules to trade.

Output
 
3945 Daily Bars 23/01/1995 Through 30/11/2010 (5790 Days)

 

Performance
Profit	5681.2447 Pts
Performance	N/A
Annualized Performance	N/A
Buy & Hold Profit	4819.6582 Pts
Buy & Hold Performance	N/A
Buy & Hold Annualized Performance	N/A

Trade Summary
Total Trades				335
Profitable Trades
Total			115
Long			64
Short			51
Average Profit			147.2962 Pts
Unprofitable Trades
Total		220
Long		103
Short		117
Average Loss		-51.1719 Pts
Accounting
Initial Equity	0.0000 Pts
Trade Profit	16939.0650 Pts
Trade Loss	-11257.8203 Pts
Commissions	784.7109 Pts
Interest Credited	0.0000 Pts
Interest Charged	0.0000 Pts
Final Equity	5681.2447 Pts

Drawdown
Highest Account Balance	5791.9947 Pts
Lowest Account Balance	-278.9460 Pts

Profit Factor: 1.5

Equity chart:

Result:

The system outperforms Nifty buy and hold by 25% over the last 15 years without much efforts.

There could be some minor variation in the results when we use Nifty futures instead of spot. 

I will dissect the trading system in later posts.

A trend defining line

Would it be not easy to have a single line define the trend. So that we go long on cross above it and then sell on cross down.

It would be more easier to follow and tradable.

Well I think there does exists such a line. But beware it is not a holy grail. It does fails many times but will make sure that you will never have that left out feeling and  can give better returns on a half yearly timeframe.

Any ideas or comments on such a line.

The back tested results link for Nifty and Bank Nifty.

System Testing Trend Defining Line

Bank Nifty: System Testing Trend Defining Line