Replies to post #38049 on A.I.M. Users Bulletin Board (AIMUSERS)

[ls7550]

Could you please identify each of the sectors next to the figures? Also, do you know how they compare with equivalent US sectors?

No sorry, that data (spreadsheet) is now gone. These figures are for quarterly figures (note july 2008 is missing) and I suspect would provide somewhat similar figures to the ones I posted earlier

Total Return 'index' figures for FT All share index and FT sectors
Date, FT All Share Index, Oil/Gas (Energy), Basic Materials, Industrial, Consumer Goods, Health Care, Consumer Services, Telecoms, Utilities, Financials, Tech

 
 1 Apr 08	3645	4770	5696	2156	4819	2715	2292	1740	4713	3856	428 
 6 Oct 08	2882	3983	3103	1674	4012	2933	1848	1438	4684	2843	378 
 31 Dec 08	2759	4763	2586	1596	4212	3232	1757	1580	4474	2070	329 
 1 Apr 09	2533	4446	2995	1479	3830	2815	1754	1398	3795	1691	386 
 1 Jul 09	2839	4716	3614	1644	4219	3005	1879	1413	4034	2138	470 
 1 Oct 09	3351	5260	4237	1966	4941	3324	2194	1651	4294	2774	566 
 31 Dec 09	3590	5749	5447	2083	5393	3588	2318	1739	4710	2691	589 
 1 Apr 10	3865	6061	6290	2338	5872	3620	2527	1814	4763	2859	688 
 1 Jul 10	3293	4415	4684	2150	5328	3483	2252	1741	4602	2472	634 
 1 Oct 10	3859	5533	5913	2464	5955	3783	2537	1969	5124	2898	781 
 30 Dec 10	4112	6038	7137	2670	6531	3721	2658	2126	5442	2905	794 
 1 Apr 11	4220	6528	7008	2760	6442	3779	2563	2260	5598	3002	866 
 1 Jul 11	4266	6304	6849	2878	6869	4141	2696	2184	5977	2965	897 
 3 Oct 11	3626	5462	4696	2385	6706	4031	2329	2169	5956	2265	813 
 30 Dec 11	3969	6491	5173	2671	7291	4416	2511	2380	5942	2349	865 
 2 Apr 12	4283	6537	5566	3083	8042	4358	2671	2421	6348	2745	985 
 2 Jul 12	4151	6211	4820	2977	8089	4459	2612	2534	6782	2627	928 
 1 Oct 12	4350	6283	5118	3195	8339	4502	2798	2553	6988	2860	1033 
 31 Dec 12	4458	5980	5417	3272	8607	4350	2967	2349	7027	3139	1142 
 2 Apr 13	4975	6359	5033	3859	10186	5088	3398	2904	7716	3502	1316 
 1 Jul 13	4908	6246	4278	3856	9831	5235	3496	3009	7701	3517	1270 
 1 Oct 13	5108	6152	4773	4158	9634	5246	3772	3437	7819	3662	1414 
 31 Dec 13	5385	6690	4851	4374	10083	5685	3960	3807	7906	3795	1507 
 1 Apr 14	5395	6662	5012	4376	10274	5973	3933	3521	8269	3737	1528 
 1 Jul 14	5517	7295	5057	4242	10741	6580	3860	3388	8672	3743	1394 
 

do you know how they compare with equivalent US sectors?

Whilst the figures would most likely differ, its the common pattern that matters.

Typically with any set of stocks/sectors that comprise a 'Index' each year you'll see most cluster quite closely around the average and a much smaller number of outliers that either make a big up (gain) or a big down (loss). Often volatility clusters i.e. a stock that drops a lot one year might rebound a lot the next year.

When you centre the figures around the average gain you eliminate the general rise (or decline) of the market and are looking purely at relative moves. i.e. sector gain / average of all sectors gain. Left as-is and those outliers induce a negative bias. Up 20% one year down -20% the next = 0.8 x 1.2 = 0.96. When you equal weighted and yearly rebalanced pairs of such assets that loss is mitigated - such that you gain by not losing so-to-speak, i.e. equal weighting the left and right tails is more inclined to yield no loss rather than a loss where no rebalancing is performed.

0.5 in a 0.8 loss
0.5 in a 1.2 gain
1.0 overall result (no overall loss). (Compared to 0.8 x 1.2 which yields a negative -4% loss).

i.e. equal weighting/rebalancing the left and right tail extremes reduces the losses that otherwise would be compounded into the total overall portfolio gain/loss.

Of course you can't predict which sectors (or stocks) will be the left or right tails in any one year in advance, so you have to equal weight all sectors (stocks) to ensure that is the case. But broadly you don't need to reset all of the set back to exact equal weightings each year and can just adjust the more extremes only (reduce the best, add to the worst). Similarly you might assume that all sectors will over the longer term tend to produce somewhat similar rewards, otherwise investors would just invest in the best single sector alone. In practice one sector will likely be the more regular winner and another the more regular loser such that over time one sector does produce a good reward and another a bad reward, but again that's unpredictable in advance.

AIM of sectors is one such method to rebalance such patterns tending to sell the most stock in the best performer, add more stock to the most worst performer. For instance FT All Share gains 2008 onwards has annualised around 6.8% since April 2008, whilst AIM/equal weighting yielded around 11.2% annualised, with around 2.7 average trades/year (16 buy/sell trades in total over the 6+ years).