Infosys - A study in following trends

Most of us hear these words on TV and press:

“Don’t buy now as the price is too high” or “Buy on correction” or “Buy Low and Sell High”.And many times we see a stock “never correcting” and continuing to keep its upward journey.

If you see my posts in the blog I advocate “buying high and selling higher” which to many people seems like going against the ‘conventional wisdom”.

To show how this philosophy actually works and how it can help us, I ran one of my trading systems which is a long term trend following system on INFOSYS to explain and understand the results.

Please note that all prices used in the analysis are adjusted for splits and bonuses.

INFOSYS was available in 1995 at a price of Rs. 7.

My algorithmic trading System gave its first BUY in April 1996 at Rs. 8.75 and this was all time high for stock then! The price moved up and then went down giving an exit at around Rs 10 in June 97 after reaching an all time high of 11.25 (Yes,we are not trying to pick the top!). The gain was around 15%.

The stock again moved up again after that to give a BUY ENTRY at 11.30 in Dec 96. This time the trend was stronger and the stock continued up to Rs 50 by Oct 97. The Exit came when stock moved back to around 40-41 levels,giving a gain of almost 300% in 10 months.

The stock again continued its upward journey in March 1998 when it moved up the 50 levels and reached all time high of 51. Again you enter on a stronger trend upward at 51 and try to ride the trend. This time trend is really a big one and the stock continues to move for two years and reaches a high of 1742 in March 2000 (The height of IT boom).

The EXIT comes when the stock starts to move back to 1100 levels (a full 600 points from the top!). Still we made 22 times our money in 2 years!

Now comes the most interesting part. Instead of moving up the stock this time begins to go downhill.From 1700 to 1100 to a low of 265 in Oct 01. Now we were not caught on the wrong foot in this fall as we did not try to “pick the bottom” or “buy low” .

The stock stayed like that till Oct 03 when again a buy signal came after the stock started to move up beyond 600. The entry point was 600 but the trend didn’t develop, so there was an exit at around 590 levels. So there was small loss of around 2%. (Also avoiding the May 04 crash)

Now the Stock again moves up in July 04 giving a BUY entry at around 760 levels and trend continues till Feb 2007 at 2095.The returns was almost 300% in 2.5 years.

The current trend in INFOSYS is downward/sideways.

Now lets summarize how much we made or lost on each trade by “buying high” and “selling higher”.

Trade 1 :15%

Trade 2:+300%

Trade 3:+2100%

Trade 4:-2%

Trade 5: 300%

So whats the moral of the story?

There are couple of observations by following this approach:

a) You are able to pick infosys at Rs 8

b)You are on the “right” side of the market most of the market cycles.

c) You are able to “stay out” of bear markets.

d) Sometimes trends continue for extended periods of time because of fear and greed factor in the markets.

e) One can never tell what is the absolute the “top” and the “bottom”. A “top” is only visible when you look back in time.When the stock made the first top of 11, who could have told you that the stock will be making a top of 2400 ?

Now the bull market in India has been continuing for last 4 years and no one can “predict” how long it might continue-5 years,10 years or 25 years!

So if it continues for another 5 years, you will be riding the “full wave” till the end and get out at the right time with your profits.

You might not believe it but the same system gave an entry in UNITECH at Rs 2.20 and gave an exit at 400.

If it gets out in 5 days, you will be out with your small losses only so that you can again wait for the right opportunity.

People might have been lucky that INFOSYS came back to 1800 levels in seven years but there were many stocks which never came back!They fell from 2000 to 20 levels and just stayed there.

So just think and when you say I am “buy” and “hold” investor or I am a long term investor.

In next post I will try to discuss how a “fundamental analyst” would have dealt with the INFOSYS stock price. It might give you a different perspective about how we view markets.

We will also try to understand how so called ‘technical analysts’ manage to create so much hoopla around their craft.

Meanwhile I will look forward to your comments.

- SAGECAPITAL

You have been using trading systems since 1996 or this is a hindsight report about what could have been?

I have been using it since then.

Hmmm .... smells of Pivots (weekly?)

Great. Can you post a backtest report of this system?

What was the actual price? Did Infosys declare any Bonus before 1995 forget split.

So should we look for stocks below Rs 10?

Good to know we have Darvas in our forum

Wonder how it did not flag out in CV's systems in 1995. Don't know if he was with any fund at that time?

Regards
ken

I don't think I need to prove myself by posting it. I have my own reasons for not posting it. If anyone/all the members here do not believe my post then they are most welcome.

Thank you.

And BTW the article is just illustrating the “buy high and sell higher” concept.

The exact entry and exit strategies are part of some of the mechanical systems I use for my personal trading. I did not mean to look for stocks below 10 or any other value or I am not illustrating how to get the "Buy" or "Sell" signals.

Apologies if I have caused any confusion.

- Sage

I dont really go back that far to test systems, market conditions were too different especially execution.I wasnt actively trading at that time and neither there were any quant funds active then.

There is no question of believing or not believing, I asked for a report because that will clarify things much clearer than just % gains because % gains are hardly the criteria for systems.Another thing is, the number of entry and exits for the system you have mentioned isnt statistically significant enough without testing it on a few more stocks.The report if you can post it will be a good case study of a reliable trend following system.