Correlation does not imply Causation

Correlation does not imply Causation

A correlation between two variables does not automatically mean that the change in one variable is the cause of the change in the values of the other variable.

Causation indicates that one event is the result of the occurrence of the other event; i.e. there is a causal relationship between the two events.

As a seasonal example, just because people in the UK tend to spend more in the shops when it's cold and less when it's hot doesn't mean cold weather causes increased high-street spending.

In this analysis, I have tried to “measure” the correlation of each stock versus the Index (it is not exactly right as the Index also “contains” the behavior of the stock whose correlation is being calculated).

From this analysis (thanks to Jahidur Rahman Khan Bhai, R expert, who helped me in doing this) I tried to “establish” relation of a stock to the Index. It could be used by “seeing” which stocks move with the index and which move “OPPOSITE” to the Index.

I used LTP at end of day (Closing Price would have been better but that would have required extra effort on my existing data set; the closing price is the final price at which a security is traded on a given trading day. The closing price represents the most up-to-date valuation of a security until trading commences again on the next trading day). Data used was from 2 May to 29 Dec 2016.

Another thing is that dividend adjustments have not been taken into account. Thus, this could be a "rough" guide.

It was done based on the following scenarios:

• As many stocks are near the end of their bullish run, we need to find new “possible” bulls.
• When the current Bull Run comes near to its end, we might see a decline. As we cannot short-sell in Dhaka Stock Exchange, we need to find which stocks might rise during the upcoming bearish period.

But the conclusion is similar – it is to find stocks at low price (to lower risk) that have not risen yet.

For this, the following table is ranked starting with stocks having highest “negative” correlation with the Index. It means that quite a few of them fell when the index rose and we “might” see an opposite reaction when the Index falls!

But let us remember that “Correlation does not imply Causation”.

Happy fishing . . . 

Company	Correlation_coef
ACI	-0.86
MARICO	-0.82
GLAXOSMITH	-0.81
EMERALDOIL	-0.78
CVOPRL	-0.76
ACMELAB	-0.76
TALLUSPIN	-0.75
BATBC	-0.73
BDTHAI	-0.70
RENATA	-0.70
AMANFEED	-0.69
SUNLIFEINS	-0.68
PRAGATILIF	-0.63
ALLTEX	-0.63
DACCADYE	-0.62
AFCAGRO	-0.60
UPGDCL	-0.58
KPCL	-0.57
GOLDENSON	-0.55
EASTRNLUB	-0.51
OLYMPIC	-0.51
ICBEPMF1S1	-0.49
ACTIVEFINE	-0.48
OAL	-0.48
BANGAS	-0.48
FARCHEM	-0.47
SANDHANINS	-0.46
LINDEBD	-0.45
LIBRAINFU	-0.45
UNITEDAIR	-0.43
ZAHINTEX	-0.43
FAMILYTEX	-0.40
BENGALWTL	-0.40
DBH	-0.39
ARAMIT	-0.39
VAMLRBBF	-0.38
GEMINISEA	-0.37
PLFSL	-0.37
SQURPHARMA	-0.37
LEGACYFOOT	-0.36
ANLIMAYARN	-0.35
BXPHARMA	-0.35
BATASHOE	-0.35
PREMIERCEM	-0.34
METROSPIN	-0.33
SAIFPOWER	-0.33
ZAHEENSPIN	-0.32
PROGRESLIF	-0.31
ITC	-0.31
POPULARLIF	-0.30
SQUARETEXT	-0.29
KEYACOSMET	-0.28
POWERGRID	-0.24
KPPL	-0.23
FBFIF	-0.22
SEMLLECMF	-0.22
SAMORITA	-0.20
NPOLYMAR	-0.19
SAMATALETH	-0.16
FAREASTLIF	-0.15
FEKDIL	-0.14
KBPPWBIL	-0.14
PADMALIFE	-0.14
NATLIFEINS	-0.13
UNIQUEHRL	-0.12
BSRMLTD	-0.11
ORIONINFU	-0.10
BEXIMCO	-0.09
ICB1STNRB	-0.09
MHSML	-0.09
PRIMELIFE	-0.08
YPL	-0.04
SINOBANGLA	-0.03
DESCO	-0.03
DESHBANDHU	-0.02
RUPALIBANK	-0.02
NFML	0.01
BSRMSTEEL	0.01
MATINSPINN	0.01
TOSRIFA	0.02
BEACHHATCH	0.05
JUTESPINN	0.06
ECABLES	0.07
FUWANGCER	0.08
ACIFORMULA	0.08