Sunday, February 5, 2012

Sector relative strength

Following on from my previous post, the relative strengths of sectors for 1 month compared to the Footsie are as follows:

ALTERNATIVE ENERGY     2 -15.93
FOOD AND DRUG RETAIL   8  -6.13
MOBILE TELECOMMUNICA   4  -5.39
HEALTH CARE EQUIPMEN   6  -5.22
TOBACCO                2  -5.07
INDUSTRIAL METALS AN   5  -3.17
GAS - WATER AND MULT   6  -3.00
ELECTRICITY            6  -2.95
EQUITY INVESTMENT IN   5  -2.88
PHARMACEUTICALS AND   12  -0.38
AEROSPACE AND DEFENC   9   0.17
REAL ESTATE INVESTME  26   0.44
FIXED LINE TELECOMMU  10   0.66
NONLIFE INSURANCE     12   0.86
LEISURE GOODS          2   1.50
INDUSTRIAL TRANSPORT   7   1.60
FOOD PRODUCERS        21   1.66
TRAVEL AND LEISURE    39   2.78
REAL ESTATE INVESTME  14   3.13
SUPPORT SERVICES      57   3.69
CONSTRUCTION AND MAT  14   4.24
FINANCIAL SERVICES    37   4.69
GENERAL INDUSTRIALS    7   4.90
ELECTRONIC AND ELECT  13   5.70
OIL AND GAS PRODUCER  45   6.15
MEDIA                 26   6.16
OIL EQUIPMENT - SERV   7   6.48
BEVERAGES              6   6.73
SOFTWARE AND COMPUTE  23   6.83
INDUSTRIAL ENGINEERI  15   6.86
LIFE INSURANCE        11   7.42
GENERAL RETAILERS     27   7.73
FORESTRY AND PAPER     1   7.83
TECHNOLOGY HARDWARE   10   8.00
HOUSEHOLD GOODS AND   10   8.81
MINING                37  10.50
PERSONAL GOODS         3  13.57
CHEMICALS              8  13.68
AUTOMOBILES AND PART   1  13.79
BANKS                  7  22.61


The second column gives you the number of companies per subsector. The third column is the relative performance. As you can seem banks have had a blistering run. Things are a little mixed up - for example mining has had a good run overall, but not "industrial metals and mining"

2 comments:

Miguel B. said...

Hey Mark,

Very nice list, how did you compile it?

Unknown said...

Hi Miguel,

I obtained the data from sharelock Holmes. Relative Strength is one of the stats that it produces. The stats were compiled using a Lisp program that I wrote myself. I did have a similar program written in R a few weeks ago - the free stats app for Linux/WIN7/OSX - but I don't have R installed on my system at the moment.

Worth mentioning is that the strengths I quoted are medians, not means. I tend to use medians more than means, as they provide more robust statistics. For relative strength, though, I think that it probably isn't too important.