Serving the Quantitative Finance Community

 
User avatar
rmeenaks
Topic Author
Posts: 7
Joined: May 1st, 2006, 2:31 pm

Dividend Forecasting

May 16th, 2008, 5:40 pm

Does anyone have algorithms, papers, etc on Dividend Forecasting? I want to know how companies such as MarkIT (http://www.markit.com/information/produ ... sting.html) does their forecasting...Thanks,Ram
 
User avatar
daveangel
Posts: 5
Joined: October 20th, 2003, 4:05 pm

Dividend Forecasting

May 16th, 2008, 6:05 pm

i dont think markit do any forecasting. you can imply dividends for prices of options and forwards or get them from equity analysts
knowledge comes, wisdom lingers
 
User avatar
KackToodles
Posts: 0
Joined: August 28th, 2005, 10:46 pm

Dividend Forecasting

May 19th, 2008, 4:15 am

dividends are voted on by the board; they are discretionary so can't be forecasted. you have to read the tea leaves but the past is no guarantee of the future.
 
User avatar
nazzdack
Posts: 0
Joined: March 3rd, 2004, 9:50 am

Dividend Forecasting

May 19th, 2008, 5:42 am

You could research how closely dividend increases correlate to increases in earnings growth over longer time frames.
 
User avatar
quantmeh
Posts: 0
Joined: April 6th, 2007, 1:39 pm

Dividend Forecasting

May 19th, 2008, 5:47 am

QuoteOriginally posted by: daveangeli dont think markit do any forecasting. u r quite wrong on this one.
 
User avatar
KackToodles
Posts: 0
Joined: August 28th, 2005, 10:46 pm

Dividend Forecasting

May 19th, 2008, 5:47 am

QuoteOriginally posted by: nazzdackYou could research how closely dividend increases correlate to increases in earnings growth over longer time frames. they correlated almost 100% over 10+ year time frame, except for wildly profitable net stocks that don't pay any dividends.
 
User avatar
rmax
Posts: 374
Joined: December 8th, 2005, 9:31 am

Dividend Forecasting

May 19th, 2008, 6:26 am

Usual method is to look at historic dividend average / trader in some places can mark to that if they beleive that the dividend is going to be wildly different (i.e. certain German German Chemical companies last year).
 
User avatar
rmeenaks
Topic Author
Posts: 7
Joined: May 1st, 2006, 2:31 pm

Dividend Forecasting

May 19th, 2008, 10:57 am

I know that dividends are voted by the board, but in order to do any modeling of equity derivatives (especially dividend paying stocks), the dividend yield and any future projected dividends needs to be computed. Historic dividends are the way I was planning on doing it, but wanted to know if there was any better way...Nazzdack has an interesting method but analyzing the time series of dividends to earnings for a particular company. I'll look into that as well...Thanks,Ram
 
User avatar
daveangel
Posts: 5
Joined: October 20th, 2003, 4:05 pm

Dividend Forecasting

May 19th, 2008, 11:11 am

i apologise as it seems that markit do actually forecast dividends. to be honest if i was doign this i would look at implied divs in the market. first, this can be done from prices of options. also, there is an active market in dividend swaps.
knowledge comes, wisdom lingers
 
User avatar
bilbo1408
Posts: 0
Joined: August 3rd, 2007, 12:50 pm

Dividend Forecasting

May 19th, 2008, 2:29 pm

What's wrong with applying historical payout percentages to earnings estimates? If the dividend is stable, and not determined by payout ratios, your task is obviously much easier. The growth rate of dividends is typically found by multipling ROE by Retention Rate.
Last edited by bilbo1408 on May 18th, 2008, 10:00 pm, edited 1 time in total.
 
User avatar
rmax
Posts: 374
Joined: December 8th, 2005, 9:31 am

Dividend Forecasting

May 19th, 2008, 3:59 pm

QuoteOriginally posted by: daveangeli would look at implied divs in the market. first, this can be done from prices of options. also, there is an active market in dividend swaps.forgot the implied divs on the options -that is used too, although with less degree of success I believe.Only seen divi swaps traded in particular cases with very low volume didn't know you could mark to them...
 
User avatar
KackToodles
Posts: 0
Joined: August 28th, 2005, 10:46 pm

Dividend Forecasting

May 20th, 2008, 5:47 am

QuoteOriginally posted by: bilbo1408What's wrong with applying historical payout percentages to earnings estimates? If the dividend is stable, and not determined by payout ratios, your task is obviously much easier. The growth rate of dividends is typically found by multipling ROE by Retention Rate. if you're going to do this, why bother at all? just assume dividends are proportional to price itself. save yourself the horrrendous work of going through the long but inaccurate exercise in futility.
 
User avatar
bilbo1408
Posts: 0
Joined: August 3rd, 2007, 12:50 pm

Dividend Forecasting

May 20th, 2008, 11:47 am

Exactly my point. I don't think any algorithm will help here, but rather just some basics pertaining to analyst's earnings estimates.
 
User avatar
gelfand
Posts: 14
Joined: July 14th, 2002, 3:00 am

Dividend Forecasting

May 21st, 2008, 9:04 pm

Companies don't like to cut "regular" dividends, so they are sticky downward, and changes in dividends will be negatively skewed, because when business is bad, a company may eliminate its dividend altogether. Companies declare "special" dividends when they want to tell the market that a dividend is one-time rather than sustainable.Dividend swaps are traded by Barclays and I assume other banks, and Barclays dividend swap indices are available on Bloomberg I believe.
 
User avatar
quantmeh
Posts: 0
Joined: April 6th, 2007, 1:39 pm

Dividend Forecasting

May 23rd, 2008, 4:39 am

QuoteOriginally posted by: gelfandCompanies don't like to cut "regular" dividends, so they are sticky downwardactually, for the same reason, they don't like to increase dividends either, because if next period they cant pay the same dividend, then the stock price goes down.