Talk:Internal rate of return

From Wikipedia, the free encyclopedia

	Business and economics Portal This article is within the scope of the Business and Economics WikiProject.
B	rated as B-Class on the assessment scale
Mid	rated as mid-importance on the assessment scale

Contents 1 Alternative method to IRR and NPV 2 IRR limitations 3 dollar-weighted return 4 external Weblink 5 Calculation of IRR 6 Cross over point method

[edit] Alternative method to IRR and NPV

The occurance of multiple IRRs and other limitations of IRR exist because of an inherent assumption in the application of the IRR method. The assumption is that the cost of borrowing and the rate of return on investment are equal. Because of this false assumption, we may get multiple IRRs or we may not get an IRR. Even if we get a single IRR, that may not be the correct IRR. A new method has been developed to solve the problem of multiple IRR without loosing the advantages of the IRR. For more information; http://www.trueirr.com/Summary%20M.htm

The relationship between the IRR and the NPV and the solution for the conflicting results are given in the above website. Financial theorists argue that the conflicting results occur due to the ‘reinvestment assumption’ implicit in all methods using the discounted cash flow technique. It has been found that the IRR method simply involves finding a compounded rate of interest and ‘reinvestment assumption’ is not done at all. In the case of certain series, the NPV may oscillate between positive and negative. A new method is developed to solve this problem too. For more information; http://www.trueirr.com/Summary%20M.htm —Preceding unsigned comment added by Krishnarajathanthri (talk • contribs) 06:14, 21 October 2007 (UTC)

The paper states that "A method called marginal IRR can be used to adapt the IRR methodology to this case." However, the method known as "Return Duration", developed in 2004 is more widely used and appropriate. More information can be found through the article: http://www.findarticles.com/p/articles/mi_qa3621/is_200401/ai_n9364326 or by a simple Google search.

"interest rate" should be replaced with "discount rate"

[edit] IRR limitations

As an investment decision tool the IRR can be misleading in the case, that the money bound in an investment is verly little or changing very much. This is the situation if investments are financed with debt. The computation of the IRR from the cash flows may result a highly positive IRR. But on very little capital bound. So you are ending with an investment that bears a high risk because of the loan and a little positive cash flow. In this cases it is better to calculate the NPV, because one may recognizies the the small value compared to the high loan. Another example is stocktrading where the money is not always invested. Therfore the bound capital is changeing very much over the investment duration and the calculated IRR is too high.

An in-deep discussion of strong points and limitations of IRR probably someone writing about this should take into account.

[edit] dollar-weighted return

I see 193.30.236.72 has removed my statement that this is also known as dollar-weighted return. I wonder why, as this is a common term. See, for example, "Investments" (1999) by Sharpe, Alexander, and Bailey, page 827. If your objection to this is the US centrism of "dollar", then I propose "money-weighted return". This is to be contrasted with time-weighted return, which is just another name for the geometric mean rate of return. What do others think? Btyner 03:22, 23 August 2005 (UTC)

You are in the right - there are two types of RR's that I know of: Dollar-weighted Rate of Return, and Time-weighted Rate of Return. Dollar-weighted is just another way of saying IRR.

[edit] external Weblink

"Capital Budgeting, audio lecture with slideshow": slideshow has few information and no audio.

[edit] Calculation of IRR

I am curious about the calculation of IRRs. Specifically, in none of the sources I've seen has anyone mentioned that not only can you get multiple answers because you have to solve an n-th degree equation, but you could also end up with a complex number as the solution, and I don't see how this could make sense. How can you tell which r is the correct answer given multiple possible rs, and what does one do in the case that the solution is complex?

Both r's are correct. One may be useful. I would argue that when there are multiple answers, prudent financial players would take the lowest. Note that multiple answers only exist, if I remember correctly, when there are more than two negative cash flows (assuming this is an investment project). One can also massage the numbers carefully to check for reality - for example, if the project has a maintenance issue in year 5 that causes negative cash flows in that year, spread the maintenance costs over two years (to avoid the negative) and see roughly where the IRR comes in. Or to put this a different way, if you have up/down cash flows like this, it is an indication that the project needs to be studied more carefully.--Gregalton 16:08, 5 January 2007 (UTC)

I don't understand how they both could be. In the case of a second degree one could end up with a positive and a negative r, for example. Say, on the example from p807 of Investments by Bodie, Kane, Marcus, where it is stated that rate of return is 7.117% (the example is ), which makes perfect sense, but the same equation could also yield an answer of -309%, which makes no sense at all. --Johs 17:09, 5 January 2007 (UTC)

In a sense it makes sense, and in a sense it doesn't. The mathematical equation itself has no problem with either value, because the equation does actually 'not' represent the true meaning of IRR from a real-world perspective, only the approximated meaning. Any value below -100% would make no sense from a financial point of view (ie you can't get a 105% discount on a product that only has a 100% value unless required by law, in which case it would be a cash outflow instead of an inflow), but from a mathematical point of view, it's very much so possible because math just doesn't care about us puny earthlings130.226.173.119 (talk) 10:11, 13 February 2008 (UTC)

I think IRR is trying to define a sort of average of periodic percent returns. That doesn't give you any sort of useful answer if some periods are positive and some negative. A 50% loss in one period is not offset by a 50% gain in the following period. e.g. Start with $100, lose 50%, you have $50. Gain 50% next period and you have $75. The average return for the two periods is 0%, but the overall loss is $25, which is 25% of the original amount. Percents don't average cleanly. I think economists use natural logs. SueHay 18:38, 17 January 2007 (UTC)

[edit] Cross over point method

I'd suggest adding information regarding the cross over point method. —The preceding unsigned comment was added by 59.167.70.183 (talk) 18:56, 17 April 2007 (UTC).