Talk:Slutsky equation

German-language article
someone just fixed the German language link, and it looks like a better article. Can anyone read German and perpahs make this one nicer? Pdbailey (talk) 06:20, 28 January 2008 (UTC)

Unexplained symbols
The article contains the following unexplained symbols:
 * Dp, Dw , p , u , and w ;

moreover, it is not clear over what domains i and j are supposed to range. What is "the matrix" referred to, and what is meant by "a given element"? --Lambiam 10:31, 15 August 2008 (UTC)
 * I hope my recent edit helps. Pdbailey (talk) 16:37, 17 August 2008 (UTC)
 * It is still unexplained over what domains i and j are supposed to range. If h(p, u) is Hicksian demand, then what is hi(p, u)?
 * A consequence of the equations is that Dp h(p, u) viewed as a function of u is constant, since the last equation can be rewritten as
 * $$D_p h(p, u) = D_p x(p, w) + D_w x(p, w) x(p, w)^\top,\,$$
 * in which the r.h.s is independent of u. Is that indeed intended? --Lambiam 21:04, 23 August 2008 (UTC)
 * Lambiam, (1) both x and h are vectors. (2) the RHS is not independant of u because
 * $$h(p, u) = x(p, w),$$
 * So there is some short hand that is useful because this form allows you to know either u or w and not have to parametrize it twice as
 * $$h[p, u(p,w)] = x(p, w),$$ and
 * $$h(p, u) = x[p, w(p,u)].$$
 * hope that clears thing up. If you think some of this should be on the page, by all means, put it there.Pdbailey (talk) 23:09, 23 August 2008 (UTC)


 * This does not clear up things for me. I still have no idea what i ranges over. Years since 1970? Volume indices of La Comédie humaine? Number of grocery stores in a 50-mile radius? If I see the notation "x(p, w)" it means to me: some function x applied to a pair of arguments p and w, so then x is a function. When you say that x is a vector, do you mean a vector-valued function? The article Hicksian demand function, which is as incomprehensible as this one (what is the min being taken over? what does it mean that u has an overbar in the defn. but not in the applications? how do x and the function u(x) enter the defn.?) does not suggest or explain how the result is a vector. How does one compare two vectors for being minimal?
 * If I understand you correctly, some abbreviations are used in the article that are not explained, and w is actually a function of two arguments, the second of which (u) is itself a function taking two arguments, the second of which is itself a function ... (and so on). So if you write out the types of these functions, you get an infinite expansion of the form
 * P × (P × (P × (P × ... → B) → A) → B) → A.
 * Somehow that seems implausible. Could you give the full types of all identifiers? --Lambiam 11:25, 24 August 2008 (UTC)
 * Lambiam, these articles are stubs, and neither I nor any other economics editor with the time appears interested in fixing that. Are you? If so, perhaps I and (knock on wood) others will help you with bringing them from their current state (as articles for any by economics Ph.D. students) to something recognizable to someone who has not done the first year of an economics Ph.D. program.
 * To answer your specific questions, x is a vector, an observable. In these treatments, it is also a function, and yes, it is a vector-valued function. If you want to know what the range of x and h are, read either the respective article's lead. As far as you infinite expansion, only expand it once, you have to choose what is the endogenous/dependent variables, and what is the exogenous/independent variable, you can choose between u and w, and p must be independent in these formulations. But the baseline is to put both u and w not as functions and assume that the reader will know to expand exactly one of them.
 * If you are interested in fixing these articles up, I would suggest starting with Expenditure minimization problem and Utility maximization problem which form the basis for all of the related articles. Pdbailey (talk) 19:00, 24 August 2008 (UTC)
 * I'm generally loathe to do more than minor touch-ups on articles, unless I understand the subject matter and have reliable sources to back me up. In this case, these articles are total gobbledygook to me. Usually I have no problem in understanding at least the mathematical formulas, but here it is all as clear to me as abracadabra. --Lambiam 20:15, 24 August 2008 (UTC)

Slutsky matrix
Which one of these is the Slutsky matrix: is it Dpx or Dph? This needs to be clarified since Slutsky matrix redirects here. Richard Pinch (talk) 20:52, 16 August 2008 (UTC)
 * It is Dph. I agree this information should be added. Pdbailey (talk) 16:42, 17 August 2008 (UTC)
 * It is indeed Dph. Some people keep changing it to Dpx. So if you notice this occurring, please revert. I just changed it to Dph. 2600:4040:7AA1:1200:78AD:B186:993F:D546 (talk) 00:10, 3 October 2023 (UTC)

mistake in copying from varian
the text: "The above equation is also known as the Hicksian decomposition of demand. The left hand side of the equation represents how demand for one good changes in response to price changes for another good. The right hand side of the equation says that this change is equal to the change in demand holding expenditure fixed at w* plus the change in demand when income changes multiplied by how much income has to change to keep utility constant." has come pretty much straight out of the book by varian, but in varian it is referring to a different equation than the one given on the wikipedia page.

this confused me for a good 10 minutes, until i looked up the reference.

not sure i'm expert enough to correct it though, as i am still studying this topic! — Preceding unsigned comment added by 137.205.222.193 (talk) 16:12, 3 December 2011 (UTC)