Enter a symbol:
For Toronto: TSX:TD
FORUM
 



 
You are here :  Home > Forum
March 29, 2024, 01:33:49 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
News: Please, write to us so we can activate your new account!
 
   Home   Help Search Login Register  
Pages: [1]   Go Down
  Print  
Author Topic: How do we calculate the portfolio performances?  (Read 65244 times)
Super Stock Picker
Administrator
Hero Member
*****

Karma: 1
Posts: 476


« on: September 04, 2006, 05:26:12 PM »

Let explain you how we calculate the return of our portfolios.
 
The rule is that there is no cash position, therefore the performance of the portfolio is divided between the performance of the stocks hold in it.

Each new position is opened with a default weight of 1. And all the stocks in the portfolio have a weight that is equal to the performance recorded since on hold. For instance, if the stock's performance has been 50% since acquisition, then its weight is 1.5. If it was -20%, then its weight would be 0.8 and so on...
This gives us the amount of each stock hold in the portfolio used to calculate the performances everyday.
 
So if you have .8, .6, 1.5, 2.3, 1.3 how would we come up with the final performance for the portfolio?
 
What's left to do the  is to multiply the performance of each stock with the number based on the stock's performance (the weight) and to add all those results and divide it by the sum of the weights.
 
In numbers that would give:
Portfolio perf = (.8*perfA + .6*perfB + 1.5*perfC + 2.3*perfD + 1.3*perfE) / (.8 + .6 + 1.5 + 2.3 + 1.3)

That's the way we calculate the portfolios' performances every day and how we come up with the return numbers published on this web site.
Logged
Super Stock Picker
Administrator
Hero Member
*****

Karma: 1
Posts: 476


« Reply #1 on: November 02, 2006, 11:31:05 PM »

I did read that post but I still don't quite understand it.

Are you saying that these are completely theoretical returns using a strategy that no investor would or could ever employ?

How would your returns compare to those of an investor who say in the case of the PEG ratio portfolio, held 12 stocks and rebalanced daily or weekly to maintain equal dollar value or equal risk across all 12 stocks held?

Is it possible that an Investor who did use proper position sizing and risk management might even do better than your mechanically derived returns?

Rex R.
Logged
Super Stock Picker
Administrator
Hero Member
*****

Karma: 1
Posts: 476


« Reply #2 on: November 02, 2006, 11:32:19 PM »

Indeed, the returns we show on the web site are theoretical as there is no commission fee taken into account.
Taking fees into account is almost impossible because every investor is a different case and it would depend on the actual size of each trade and the fees charged by the broker used.
 
In order to make it simple, the choice has been to present our portfolios has a set of stocks that compares to an index (in which there is no commission fee neither).
 
But it is still possible to rebalance the portfolio every time a new order is issued. We currently propose weekly and monthly updated portfolios, so that would be the rate of the rebalancing too. If, let's say that commission fees are very low, it is reasonable to think that you could rebalance at each update.
 
Now, note that we do not propose to rebalance the portfolio to maintain equal dollar value. We propose to rebalance in order to get the capital fully invested at any time. Then, if you end up with the same ratio of the portfolio invested in each stock as we propose, you will get returns that will be the same way (up or down) each day, but only, as your cash position is not moving, you will have less volatility when the portfolio rises or decreases.
If you end up changing the ratio of stocks hold, then it is impossible to say as it will depend if you give more or less weight to a performing stock or to a dragging one.
 
So, it is for sure possible to get even better that what we show. The first idea that would be more profitable is to be 100% invested when the market rises and less when it falls. This way, you can profit of the good markets and be less exposed to the bad ones.
 
How you chose where to stand, that we cannot answer...
Logged
Victor
Sr. Member
****

Karma: 1
Posts: 73


« Reply #3 on: December 02, 2007, 12:39:59 AM »

Just to be clear:
1) the posted results reflect daily rebalancing.
2) the weights used are yesterday's performance percentages times today's performance.

Can you confirm this?

Just want to ensure it's not today's performance percentage (since purchase) times today's performance.

thanks,

Victor
Logged
Victor
Sr. Member
****

Karma: 1
Posts: 73


« Reply #4 on: December 02, 2007, 01:01:25 AM »

I just saw this in a news release:

"Super Stock Picker's portfolio performance figures are based on the total return of non-rebalanced simulated portfolios, transaction fees not included. Returns reflect the reinvestment of dividends and other earnings. "

Are you keeping track of non-rebalanced portfolios as well?

Not sure I'm following how this works ( the rebalancing part ).

Thanks,

Victor

Logged
Super Stock Picker
Administrator
Hero Member
*****

Karma: 1
Posts: 476


« Reply #5 on: December 03, 2007, 03:26:35 PM »

Just to be clear:
1) the posted results reflect daily rebalancing.
2) the weights used are yesterday's performance percentages times today's performance.

Can you confirm this?
Hello Victor,

The posted results reflect the case when all the money is always fully invested and there is no cash position in the portfolio.

So, that implies to rebalance the portfolio every time there is a new buy or a new sell.

Daily, there is nothing done on our side, the only thing that will change the weights in the portfolio, it is the performances of each stock.

Here is an example:
You hold two stocks, which have not moved since you had bought them.
In that case, the money in the portfolio is divided equally between the two positions.
On a given day, one stock moves up 10%, and the other one down 10%. Of course, the weights of those two stocks in the portfolio have changed, but I don't think that you should call this rebalancing. That's what happen everyday in any portfolio...

Now, if you would buy a new position, then, you need to reallocate your money. That's when we rebalance the position, weighting each of them proportionally to their performance since they were recommended in the portfolio.
Logged
Super Stock Picker
Administrator
Hero Member
*****

Karma: 1
Posts: 476


« Reply #6 on: December 03, 2007, 03:32:14 PM »

I just saw this in a news release:

"Super Stock Picker's portfolio performance figures are based on the total return of non-rebalanced simulated portfolios, transaction fees not included. Returns reflect the reinvestment of dividends and other earnings. "

Are you keeping track of non-rebalanced portfolios as well?

Not sure I'm following how this works ( the rebalancing part ).

Thanks,

Victor
Hello again,

What we mean by non-rebalanced in our press releases, is that we do not rebalance the positions every so often (let's say weekly), to maintain an equally dollar weighted portfolio as people are more used to see.

Rebalancing happens only when we add or remove a stock in the portfolio, and we choose to let the big winners run into very large position in the portfolio.

You should have a look at the virtual $50,000 portfolio as you'll see weekly examples of this mechanism.
Logged
Victor
Sr. Member
****

Karma: 1
Posts: 73


« Reply #7 on: December 05, 2007, 06:34:42 PM »

Ok, so rebalancing would not occur on a week when there is no recommendation (buy or sell).  To the extent there's changes every week (typical of the PMW system), the rebalancing occurs "weekly."  For the V5 system the rebalancing is much more sporadic.

I think I've got it now.

Can you confirm the above?

Thanks,

Victor
Logged
Super Stock Picker
Administrator
Hero Member
*****

Karma: 1
Posts: 476


« Reply #8 on: December 06, 2007, 10:56:20 AM »

Hello Victor,

Yes, you got it all right. That's the way rebalancing happens.
Logged
jjcthorpe
Newbie
*

Karma: 0
Posts: 1


« Reply #9 on: April 14, 2008, 11:30:58 AM »

How does that math work out?

If I had two stocks in a portfolio that were bought on the same day, equal amount of money in each initially, and one went up 50% and the other went down 50%, the total performance would be 0(zero) correct?

i.e  $ 100 in each stock initially ( $200 total investment), one loses $50 ( -50% performance) and one gains $50 ( + 50% performance), the total performance of the portfolio is 0 ( or 0%).

By your calculations , weight x performance, the portfolio performance would be 25%:

[(0.5)(-50%) + 1.5 (+50%)]/ (0.5 + 1.5) = [-25% + 75%]/2 = 50%/2 = 25%.

Your calculations minimize the effects of the losers and inflate the effects of the winners doesn't it? i.e a stock that lost 100% would have no effect ( rather than a negative effect) on the performance because the weight would now be 0. Maybe I'm confused. Let me know.

Thanks.
Logged
Super Stock Picker
Administrator
Hero Member
*****

Karma: 1
Posts: 476


« Reply #10 on: April 14, 2008, 09:46:23 PM »

If I had two stocks in a portfolio that were bought on the same day, equal amount of money in each initially, and one went up 50% and the other went down 50%, the total performance would be 0(zero) correct?
Yes, that's correct.

By your calculations , weight x performance, the portfolio performance would be 25%:

[(0.5)(-50%) + 1.5 (+50%)]/ (0.5 + 1.5) = [-25% + 75%]/2 = 50%/2 = 25%.
What is wrong in this calculation, is that you applied the wrong weights.
In the above formula, you are calculating the performance of a portfolio in which the two stocks have the same weight, 1, as they are fresh buys, with a performances of 0% each since on hold.
So, to correct it:
[(1)(-50%) + 1 (+50%)]/ (1 + 1) = [-50% + 50%]/2 = 0%/2 = 0%.

Now, the weights you have used would be the correct ones for the next day...

Your calculations minimize the effects of the losers and inflate the effects of the winners doesn't it? i.e a stock that lost 100% would have no effect ( rather than a negative effect) on the performance because the weight would now be 0. Maybe I'm confused. Let me know.
Our calculation reflects the way we allocate money in our portfolios.
Indeed, we overweight the good performers and underweight the bad performers.
Also, a stock that lost 100% would indeed be wiped out of the portfolio, but not without hitting hard the portfolio's performance that day...
Logged
ruth
Jr. Member
**

Karma: 0
Posts: 8


« Reply #11 on: December 22, 2008, 03:14:22 PM »

Dear SSP administrator

I am trying to re-balance the SSP v.4 portfolio to match your returns.

I am fine through Dec 1, 2008:   -46.57% return which uses the correct WN Nov 3 purchase price of 60.98

In order to get the same return on Dec 2nd,  I have to re-balance the portfolio as if WN had your Nov 3rd purchase price of 60.63.

Then things don’t work using either pricing for WN (i.e. 60.98 the correct end of day price or the SSP quoted price of 60.63) … please tell me what your allocation percentages are for Dec 8, 15, 19.  If you don’t have time for all 3,even getting one of them will probably put me on the right track again.

The following are my Dec 15 allocations based on WN purchase price of 60.98

WN            0.934732699
FFH          .101552795
WPK        1.059375
NWF.UN            1
SXC                  1

Matching your returns is important to how I rebalance the SSP v.4 portfolio

Your response is much appreciated
Ruth
Logged
Super Stock Picker
Administrator
Hero Member
*****

Karma: 1
Posts: 476


« Reply #12 on: December 22, 2008, 11:03:53 PM »

Hello Ruth,

For the calculation of the today's performances and to allocate your portfolio the same way as our models, you just need to take into account the performances of the recommendations as shown on the portfolio's description page.

Now, about the problem you have seen, I think it is because WN has recently recorded a dividend payment.

As the result, on the day it is published in our database, we subtract that dividend from the price actually paid. If you look on the historical orders page, you can see this "adjusted" buy price. We also account this dividend in the performance of that stock for the payment day.

Look on this page to see the close prices and the adjusted close prices of WN:
http://finance.yahoo.com/q/hp?s=WN.TO

Feel free to ask more questions if you need.
Logged
ruth
Jr. Member
**

Karma: 0
Posts: 8


« Reply #13 on: December 23, 2008, 10:17:06 AM »

Thank you so much.  It makes sense now. 
And thanks for the adjusted prices site.
Ruth

Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Valid XHTML 1.0! Valid CSS!
 
 
 
Copyright ©2004-2023 Agnosoft