Marginal effects significance vs?original model effects significance - Statalist?

關(guān)于IVprobit模型邊際效應(yīng)顯著性與原始模型效應(yīng)顯著性對比：statalist

https://www.statalist.org/forums/forum/general-stata-discussion/general/1329201-marginal-effects-significance-vs-original-model-effects-significance

Richard Williams:（貼主）

I often get questions like "This variable has significant effects in the original (logit/probit/heckprobit/whatever) model. But its marginal effect is not significant. Why?" Or vice-versa. I tend to just say that different hypotheses are being tested. But can somebody give a more elegant or complete explanation?

我經(jīng)常會遇到這樣的問題："這個變量在原始（logit/probit/heckprobit/什么的）模型中效果顯著。但其邊際效應(yīng)并不顯著。為什么？或者反之亦然。我傾向于只說正在檢驗不同的假設(shè)。但有人能給出更優(yōu)雅或更完整的解釋嗎？

Clyde Schechter:

I think this is a multi-faceted problem and that different misunderstandings underlie this question when asked by different people. Among them:

我認(rèn)為這是一個多方面的問題，不同的人在提出這個問題時會產(chǎn)生不同的誤解。其中包括

1. Some are making a fetish out of 0.05. They see a p-value of 0.04 in one place and 0.06 in another where they expect them to be "the same" and they freak out.

1.有些人迷信 0.05。他們看到一個地方的 p 值是 0.04，而另一個地方的 p 值是 0.06，他們希望兩者 "一樣"，于是他們就“抓狂”了。

2.Some have a fundamental misconception of what statistical significance is. They think that statistically significant means "there is an effect" and not statistically significant means "there is no effect." Which again leads them to freak out when they get these apparently contradictory results. For these people, what is needed is re-education about the concept of statistical significance. They need to learn that "statistical significance" is an arbitrary dichotomization of a continuum of degrees of improbability of the outcome under the null hypothesis. I try to get them to think of it differently: a statistically significant result means a combination of the effect in the sample being large enough, and the data being of adequate quantity and quality (low noise) that our estimates of the effect have enough precision that we are highly skeptical of the idea that the true effect size is zero. So our p-values are telling us only indirectly about how big the effect is, and not even actually telling us whether or not it really is zero.?

2. 有些人對什么是統(tǒng)計學(xué)意義有根本性的誤解。他們認(rèn)為統(tǒng)計意義顯著意味著 "有影響"，而統(tǒng)計意義不顯著意味著 "沒有影響"。這再次導(dǎo)致他們在得到這些明顯矛盾的結(jié)果時抓狂。對于這些人來說，需要的是對統(tǒng)計學(xué)意義概念的再教育。他們需要知道，"統(tǒng)計顯著性 "是對零假設(shè)下結(jié)果不可能程度連續(xù)體的任意二分法。我試圖讓他們換個角度思考：統(tǒng)計意義上的結(jié)果意味著樣本中的效應(yīng)足夠大，而且數(shù)據(jù)的數(shù)量和質(zhì)量足夠好（噪聲低），我們對效應(yīng)的估計有足夠的精確度，以至于我們對真實效應(yīng)大小為零的想法持高度懷疑態(tài)度。因此，我們的 p 值只能間接地告訴我們效應(yīng)有多大，甚至不能告訴我們效應(yīng)是否真的為零。

I generally prefer to focus them on the confidence intervals, because those are more conducive to thinking about an estimate and the precision of that estimate. Then the p-value can be understood as telling us whether our estimate is both sufficiently far from zero and sufficiently price that it is implausible that we would get such an estimate from a truly zero effect. I also like to point out that in most real-life research situations, the null hypothesis of zero effect is a rather shabby straw man in any case. (As you can tell, I"m not a big fan of p-values.)

一般來說，我更愿意把重點(diǎn)放在置信區(qū)間上，因為置信區(qū)間更有利于思考估計值和估計值的精確度。那么，P 值就可以理解為告訴我們，我們的估計值是否離零足夠遠(yuǎn)，價格是否足夠高，以至于我們無法從真正的零效應(yīng)中得到這樣的估計值。我還想指出，在現(xiàn)實生活中的大多數(shù)研究中，零效應(yīng)的零假設(shè)無論如何都是一個相當(dāng)寒磣的稻草人。(正如你所知道的，我不太喜歡 p 值）。

As an antidote to #1 and #2 I often advise those I mentor not to use the term "statistically significant" in my presence.

The above two are generic and give rise to a lot of misunderstandings and confusion about results from a wide variety of analyses.

作為#1和#2的解藥，我經(jīng)常建議我的導(dǎo)師不要在我面前使用“統(tǒng)計上顯著”這個詞。

上面兩個是通用的，從各種各樣的分析結(jié)果引起了很多誤解和混亂。

More specific to the situations you mention are the following:

更具體到你提到的情況有以下幾種：

3. Failure to take into account that the coefficients and odds ratios are different metrics from the marginal effect on probability. The non-linearity of these models then produces "paradoxical" results. In fact, in many situations, if you run -margins, dydx(x) at(x = (numlist spanning a wide range of values)) you will get an interesting mix of large and small, "significant" and "non-significant" marginal effects. Drawing a graph of the logistic curve and pointing out that it has a steep section in the middle and flat sections at the ends probably makes the point better than any number of words and sentences. So a unit increment in x may correspond to a rather large increase in predicted outcome probability if we are starting out in the steep area, and a barely visible increase if we are out at the far ends. Once again, due to the noise in our data and sampling variation, we are estimating these marginal effects with a certain degree of precision, and some, but not all, marginal effects will be large enough that we can bound them away from zero at that degree of precision.

3. 沒有考慮到系數(shù)和幾率比是不同于概率邊際效應(yīng)的指標(biāo)。這些模型的非線性就會產(chǎn)生 "自相矛盾 "的結(jié)果。事實上，在很多情況下，如果運(yùn)行-margins, dydx(x) at(x = (numlist spanning a wide range of values))，你會得到一個有趣的混合結(jié)果，即邊際效應(yīng)有大有小，有 "顯著 "的，也有 "不顯著 "的。畫一張邏輯曲線圖，指出它的中間部分比較陡峭，兩端比較平坦，可能比任何文字和句子都更能說明問題。因此，如果我們從陡峭區(qū)域開始，x 的單位增量可能對應(yīng)于預(yù)測結(jié)果概率的大幅增加，而如果我們在遠(yuǎn)端，則幾乎看不到增加。同樣，由于數(shù)據(jù)中的噪聲和抽樣變化，我們在估算這些邊際效應(yīng)時會有一定的精確度，有些邊際效應(yīng)會大到足以讓我們在這種精確度下將其從零束縛開來，但并非所有邊際效應(yīng)都是如此。

?Just where we are on the logistic curve isn't always obvious from looking at the regression output or the marginal effects, as it depends on the sample distributions of the predictor variables too. The predicted probability for the sample as a whole, or with all variables set at their means, can be helpful for figuring that out. Once you know where you are on the curve, it is easier to see graphically why a marginal effect might be surprisingly large or small in the face of a particular logistic regression coefficient.

從回歸結(jié)果或邊際效應(yīng)來看，我們在邏輯曲線上的位置并不總是很明顯，因為這也取決于預(yù)測變量的樣本分布。樣本整體的預(yù)測概率，或?qū)⑺凶兞吭O(shè)置為均值時的預(yù)測概率，可以幫助我們弄清這一點(diǎn)。一旦知道自己在曲線上的位置，就更容易從圖形上看出為什么邊際效應(yīng)在特定的邏輯回歸系數(shù)面前會出奇地大或小。

Stephen Jenkins:

+1 to the post from Clyde. Very nicely put.
My tuppenceworth worth relates to Rich's statement that:I tend to just say that different hypotheses are being tested.

+1克萊德。說得非常好。

我的一點(diǎn)看法與里奇的發(fā)言有關(guān)，他說：我傾向于只說正在測試不同的假設(shè)。

Marginal effects are typically (but not always) non-linear functions of all the estimated parameters and explanatory variables. So even if particular coefficient or OR is "statistically significant", it doesn't guarantee that the marginal effect associated with that coefficient is "statistically significant". For that reason, you can get some of the features described by Clyde in #2.

邊際效應(yīng)通常（但不總是）是所有估計參數(shù)和解釋變量的非線性函數(shù)。因此，即使特定系數(shù)或 OR 具有 "統(tǒng)計顯著性"，也不能保證與該系數(shù)相關(guān)的邊際效應(yīng)具有 "統(tǒng)計顯著性"。因此，你可以得到克萊德在 #2 中描述的一些特征。

Chandra Shah:

Clyde's first point is trivial and that is not the point of the question.The question is not really about why odds ratio is "significant" while the corresponding marginal effect is not. It is really about what does it mean when the respective p-values are very 'far' apart (e.g. the p-value for the odds ratio is .002 an that for the marginal effect is .5). Is this plausible and under what conditions? I accept Stephen's point that the calculation of the marginal effects are typically non-linear functions of all the estimated parameters and explanatory variables. But does this always mean less precision in its calculation? Under what conditions would you get the opposite result? Certainly if you get such large non-trivial differences in the p-values then it is important ask the question about could possibly be the cause.

克萊德的第一點(diǎn)微不足道，這不是問題的重點(diǎn)。實際上，問題并不在于為什么幾率比是 "顯著的"，而相應(yīng)的邊際效應(yīng)卻不顯著。真正的問題是，當(dāng)各自的 p 值相差很 "遠(yuǎn) "時（例如，幾率比例的 p 值是 0.002，而邊際效應(yīng)的 p 值是 0.5），這意味著什么？這合理嗎？我接受斯蒂芬的觀點(diǎn)，即邊際效應(yīng)的計算通常是所有估計參數(shù)和解釋變量的非線性函數(shù)。但這是否總是意味著計算精度較低？在什么情況下會得到相反的結(jié)果？當(dāng)然，如果您得到的 p 值差異如此之大，那么就有必要問一問可能的原因是什么。

Stephen Jenkins：

I would agree with Chandra's remark that:

Certainly if you get such large non-trivial differences in the p-values then it is important ask the question about could possibly be the cause.

我同意錢德拉的觀點(diǎn)

當(dāng)然，如果 p 值出現(xiàn)如此大的非微小差異，那么就有必要提出一個問題：原因可能是什么？

However, I doubt that it's possible to set out, in a general way, the conditions under which "large" divergences appear. Won't it be model-specific? Also, it's likely to be related to how non-linear is the function relating the original statistic of interest and the marginal effect of interest. (And remember that there is no single marginal effect in a non-linear model, so "which marginal effect?" is something for Richard as well.) In this respect, Clyde's point 3 is definitely relevant here.

然而，我懷疑是否有可能籠統(tǒng)地列出出現(xiàn) "大 "分歧的條件。這難道不是因模型而異嗎？而且，這很可能與相關(guān)原始統(tǒng)計量和相關(guān)邊際效應(yīng)之間的非線性函數(shù)有關(guān)。(請記住，在非線性模型中不存在單一的邊際效應(yīng)，因此 "哪個邊際效應(yīng)？"也是理查德需要考慮的問題）。在這方面，克萊德的第 3 點(diǎn)無疑與此有關(guān)。

Richard Williams：

Excellent points, and I especially like how Clyde explains point 3. Different hyp are being tested, And, as Stephen adds, there is no single marginal effect. The value of the marginal effect is contingent on how the values of the other variables in the model are set. You may use atmeans or asobserved and get a single number but there are any number of other ways you could set the values of the other variables.

說得非常好，我尤其喜歡克萊德對第 3 點(diǎn)的解釋。正如斯蒂芬所補(bǔ)充的，沒有單一的邊際效應(yīng)。邊際效應(yīng)的值取決于模型中其他變量值的設(shè)定。您可以使用 atmeans 或 asobserved 得到一個單一的數(shù)字，但您也可以通過其他多種方式設(shè)置其他變量的值。

Personally, I mostly focus on the significance of coefficients, not the significance of the marginal effect. Or, if I do look at the significance of the marginal effect, it might be over a range of values. So, for example, the marginal effect of race might be very small at low values of age but much greater at higher values of age. e.g. something like

就我個人而言，我主要關(guān)注系數(shù)的顯著性，而不是邊際效應(yīng)的顯著性?；蛘哒f，如果我真的要看邊際效應(yīng)的顯著性，那也可能是在一個數(shù)值范圍內(nèi)。例如，在年齡值較低時，種族的邊際效應(yīng)可能非常小，但在年齡值較高時，種族的邊際效應(yīng)就會大得多。

Code:

webuse nhanes2f, clear?

svy: logit diabetes i.black i.female age, nolog?

margins, dydx(black) at(age=(20 30 40 50 60 70)) vsquish?

marginsplot

Chandra Shah：

Quote from Greene:

An empirical conundrum can arise when doing inference about partial effects rather than coefficients. For any particular variable, wk, the preceding theory does not guarantee that both the estimated coefficient, θk and the associated partial effect, δk will both be ‘statistically significant,’ or statistically insignificant. In the event of a conflict, one is left with the uncomfortable problem of simultaneously rejecting and not rejecting the hypothesis that a variable should appear in the model. Opinions differ on how to proceed. Arguably, the inference should be about θk, not δk, since in the latter case, one is testing a hypothesis about a function of all the coefficients, not just the one of interest.

引自格林：

在推斷部分效應(yīng)而非系數(shù)時，可能會出現(xiàn)一個經(jīng)驗難題。對于任何特定變量 wk，前面的理論并不能保證估計系數(shù) θk 和相關(guān)的部分效應(yīng) δk 都 "在統(tǒng)計上顯著 "或 "在統(tǒng)計上不顯著"。在出現(xiàn)沖突的情況下，我們就會面臨一個令人不安的問題，即同時拒絕和不拒絕某個變量應(yīng)出現(xiàn)在模型中的假設(shè)。對于如何處理這個問題，眾說紛紜。可以說，推論的對象應(yīng)該是 θk，而不是 δk，因為在后一種情況下，我們是在檢驗關(guān)于所有系數(shù)的函數(shù)的假設(shè)，而不僅僅是感興趣的系數(shù)。

And I suppose in the case of a bivariate probit with sample selection, the calculation of δk is much more complicated involving parameter estimates and variables from the selection equation as well.

而且我認(rèn)為，在有樣本選擇的二元 probit 的情況下，δk 的計算要復(fù)雜得多，其中還涉及參數(shù)估計和選擇方程中的變量。

Does this mean you are more likely to encounter such conundrums when you have a more complex model like a bivariate probit with sample selection than a simple probit?

Is this anything to do with the consistency of the estimator for the variance-covariance matrix?Or the delta method used in the calculation?

這是否意味著，與簡單的 probit 模型相比，在使用帶有樣本選擇的雙變量 probit 這樣更復(fù)雜的模型時，更容易遇到這樣的難題？

這是否與方差-協(xié)方差矩陣估計器的一致性有關(guān)？還是計算中使用的 delta 方法？

Stephen Jenkins：

Please provide an exact and full bibliographic citation for the quotation from Greene.
I would suggest that the issues are primarily to do with how non-linear the transformations are that relate the original parameter(s) to the partial/marginal effects. (And not the issues raised in your last 2 questions.)

請?zhí)峁┮?Greene 的確切、完整的參考文獻(xiàn)。

我認(rèn)為，這些問題主要與將原始參數(shù)與部分/邊際效應(yīng)聯(lián)系起來的變換的非線性程度有關(guān)。(而不是你在最后兩個問題中提出的問題）。

Marcos Almeida：

I kindly ask you to provide quotes, well, under quotation marks (as shown above), as recommended in the FAQ. By reading #7, it turns out difficult to many of us to spotlight who said who, I mean, message and unmarked quotation, unfortunately, got entwined.

我懇請您按照常見問題的建議，在引號下（如上圖所示）提供引文。通過閱讀 #7，我們中的很多人都很難發(fā)現(xiàn)是誰說了誰的話，我的意思是，不幸的是，信息和未加標(biāo)記的引文糾纏在了一起。

Chandra Shah：（回應(yīng)引文）

page 12:?http://archive.nyu.edu/bitstream/2451/26036/2/7-7.pdf
Also, see:
Dowd, BE, Greene, WH & Norton, EC 2014, 'Computation of Standard Errors', Health Services Research, vol. 49, pp. 731-750.
where the authors discuss the issue of consistency of the estimator of variance- covariance matrix and the complexity involved in calculating the standard error of the a marginal effect in a multiple equation model (e.g. bivariate probit with sample selection). If full information maximum likelihood is used then the estimator is consistent.
Sorry I can't figure out how to do the quotes like in the above posts!

第 12 頁：http://archive.nyu.edu/bitstream/2451/26036/2/7-7.pdf

另見，

Dowd, BE, Greene, WH & Norton, EC 2014, 'Computation of Standard Errors', Health Services Research, vol. 49, pp.

作者在書中討論了方差-協(xié)方差矩陣估計值的一致性問題，以及計算多方程模型（如帶有樣本選擇的雙變量 probit）中邊際效應(yīng)標(biāo)準(zhǔn)誤差的復(fù)雜性。如果使用全信息最大似然法，那么估計值是一致的。

抱歉，我不知道如何像上面的帖子那樣使用引號！

Stephen Jenkins：

Chandra: please read the FAQ from top to bottom. (Hit the black bar at the top of the page.) Your post #11 suggests you have mastered how to do the "quote" inserts, but also read about using CODE delimiters. Also read how to use the Advanced editor and its functionality (accessed by clicking on the underlined upper-case A in the editing box for composing messages). With that you can insert hyperlinked URLs

Chandra: 請從上到下閱讀常見問題（點(diǎn)擊頁面頂部的黑條）。你的帖子 #11 表明你已經(jīng)掌握了如何插入 "引用"，但也請閱讀有關(guān)使用 CODE 分隔符的內(nèi)容。還請閱讀如何使用高級編輯器及其功能（點(diǎn)擊編輯框中的下劃線大寫 A 即可進(jìn)入）。有了它，你可以插入超鏈接 URL

For Stata's capabilities for estimating marginal effects for a "bivariate probit with sample selection", see?help heckprobit_postestimation.?Also read the associated manual entry for information about methods and formulae. The "complexity" you refer to exists of course, but StataCorp have done a lot of work for you.

關(guān)于 Stata 估算 "帶樣本選擇的雙變量 probit "邊際效應(yīng)的功能，請參閱 help heckprobit_postestimation。還可以閱讀相關(guān)的手冊條目，了解有關(guān)方法和公式的信息。您所說的 "復(fù)雜性 " (δk 的計算要復(fù)雜得多，其中還涉及參數(shù)估計和選擇方程中的變量) 當(dāng)然存在，但 StataCorp 已經(jīng)為您做了大量工作。

Chandra Shah：

Thanks Stephen, just found the Advanced editor!

謝謝你，斯蒂芬，我剛找到高級編輯器！

I have used Stata capabilities you refer to for estimating my bivariate model with sample selection. Both the outcome and selection equations are important for the research questions I am investigating. Furthermore I have replicate weights, plausible values and missing data to deal with. For each plausible value I have three multiple imputed data sets, making 30 multiple imputed data. Incidentally the two variables for which the p-values of the coeff. and the marginal effect are radically 'far' apart are both continuous.

And thanks everybody for your contribution to this thread.

我使用了您提到的 Stata 功能來估計帶有樣本選擇的雙變量模型。結(jié)果方程和選擇方程對我正在研究的問題都很重要。此外，我還需要處理重復(fù)權(quán)重、可信值和缺失數(shù)據(jù)。對于每個似然值，我都有三個多重估算數(shù)據(jù)集，即 30 個多重估算數(shù)據(jù)。順便提一下，系數(shù)和邊際效應(yīng)的 p 值相差很遠(yuǎn)的兩個變量都是連續(xù)變量。

感謝大家對本主題的貢獻(xiàn)。

Richard Williams：

That kind of complexity makes my head hurt! I might try skipping a few of the bells and whistles (e.g. do it without the imputation, or without the replicate weights) and see if that changes anything. I wouldn't expect it to, but a mistake somewhere along the way (either by you or by Stata) might affect the results.

這種復(fù)雜性讓我頭疼！我可能會試著跳過一些繁瑣的程序（例如，不進(jìn)行估算，或不使用重復(fù)權(quán)重），看看是否會有什么變化。我不指望會有什么變化，但過程中的某個錯誤（無論是你還是 Stata）可能會影響結(jié)果。

Sebastian Kripfganz：

From a technical perspective, Stephen's answer is already sufficient. Coefficients and marginal effects are different quantities, and the latter are often non-linear combinations of the former. Typically, one would expect the p-values to be similar but there is no reason why they have to be close or even identical.

從技術(shù)角度看，斯蒂芬的回答已經(jīng)足夠了(將原始參數(shù)與部分/邊際效應(yīng)聯(lián)系起來的變換的非線性程度)。系數(shù)和邊際效應(yīng)是不同的量，后者往往是前者的非線性組合。通常情況下，我們會希望 p 值相似，但沒有理由說它們必須接近甚至相同。

With regard to the discussion whether to look at coefficients or p-values, we should ask: Why are we interested in the respective test result? The coefficients themselves in non-linear models typically do not have a good interpretation when we want to examine the effect of a certain covariate on our outcome variable. When interpreting the results, we would typically look at the marginal effects. Yet, testing for statistical significance of the coefficient estimates can be relevant when we think about the model specification, whether to include or exclude a certain covariate.

關(guān)于是看系數(shù)還是看 p 值的討論，我們應(yīng)該問：為什么我們對各自的檢驗結(jié)果感興趣？當(dāng)我們想研究某個協(xié)變量對結(jié)果變量的影響時，非線性模型中的系數(shù)本身通常沒有很好的解釋。在解釋結(jié)果時，我們通常會關(guān)注邊際效應(yīng)。然而，當(dāng)我們考慮模型的規(guī)格時，檢驗系數(shù)估計值的統(tǒng)計顯著性可能與我們的考慮相關(guān)，即是否包含或排除某個協(xié)變量。

A variable could be statistically relevant in the sense that its coefficient estimate is statistically significant but at the same time its marginal effect might be statistically insignificant. The latter does not mean that this variable does not matter because it also affects the marginal effects of all the other covariates. In other words, including it still improves the fit of the model.

一個變量在統(tǒng)計意義上可能是相關(guān)的，即其系數(shù)估計值在統(tǒng)計意義上是顯著的，但同時其邊際效應(yīng)在統(tǒng)計意義上可能是不顯著的。后者并不意味著該變量不重要，因為它也會影響所有其他協(xié)變量的邊際效應(yīng)。換句話說，加入該變量仍能改善模型的擬合效果。

Chandra Shah：

Thanks for the latest comments.

As you rightly point out Richard, the results are in the same ball park without multiple imputation and replicate weights. But I only know this after having estimated the models both ways! There were 10% missing values in aggregate across all variables. As Paul Allison noted, one should impute missing values whenever possible. Also the use of replicate weights and plausible values is what the OECD recommend for the analysis of the PIAAC data that I am using.

感謝您的最新評論。

正如您正確指出的那樣，理查德（這種復(fù)雜性讓我頭痛?。?，在沒有多重估算和重復(fù)權(quán)重的情況下，結(jié)果是相同的。但我是在對兩種方法的模型都進(jìn)行了估計之后才知道這一點(diǎn)的！所有變量的缺失值合計為 10%。正如保羅-埃里森（Paul Allison）所指出的，只要有可能，就應(yīng)該對缺失值進(jìn)行估算。此外，經(jīng)合組織也建議在分析我正在使用的 PIAAC 數(shù)據(jù)時使用重復(fù)權(quán)重和可信值。

Elise Sobrie：（also我的疑問）

Dear all

After reading a couple of times your posts, I understand what you are saying. However, it is not clear to me whether to follow the p-values of the margins or those from the coefficients in the Original model. I know different hypotheses etc. are beign tested, but just one clear answer which to follow?

Thank you in advance!

親愛的各位

讀了幾遍你們的帖子后，我明白了你們的意思。但是，我不清楚是應(yīng)該遵循邊際的 p 值，還是遵循原始模型中系數(shù)的 p 值。我知道要對不同的假設(shè)等進(jìn)行檢驗，但只想得到一個明確的答案，到底應(yīng)該遵循哪個？

在此先表示感謝！

Richard Williams：（回應(yīng)Elise）

As I already stated, my preference is to use the p-values from the original coefficients. I think that is also what is more common. Many analyses do not even present adjusted predictions or marginal effects, but you just about always have the coefficients.

But I am not the ultimate authority on such things, so you should decide what is best for you given what you want to test.

我已經(jīng)說過，我更傾向于使用原始系數(shù)的 p 值。我認(rèn)為這也是更常見的做法。很多分析甚至沒有提出調(diào)整后的預(yù)測值或邊際效應(yīng)，但你幾乎總是有系數(shù)。

但我不是這方面的終極權(quán)威，所以您應(yīng)該根據(jù)自己想要測試的內(nèi)容來決定哪種方法最適合您。

Evelyn Mare：

I have encountered a similar problem. My logit models suggests a p value of <.05 and my logit model a p value >.15.

Originally posted by?Richard Williams?View Post

As I already stated, my preference is to use the p-values from the original coefficients. I think that is also what is more common

我也遇到了類似的問題。我的 logit 模型顯示 p 值小于 0.05，而我的 logit 模型顯示 p 值大于 0.15。“理查德：我已經(jīng)說過，我傾向于使用原始系數(shù)的 p 值。我認(rèn)為這也是更常見的做法”

I would lean towards presenting the AMEs, because they are easier to interpret, yet I am wrangling with the insignificance.

我傾向于提出 AMEs，因為它們更容易解釋，但我也在糾結(jié)其微不足道之處。

Edmondo Ricci：

Dear Clyde,
Thank you for your post. Since tests on marginal effect and coefficients are different tests, I can see statistical significance can be different.
But would it be even possible that coefficients are positive and marginal effect is negative?
I managed to make a dataset where I get following results.

親愛的克萊德

感謝您的來信。由于邊際效應(yīng)檢驗和系數(shù)檢驗是不同的檢驗方法，我知道統(tǒng)計意義可能不同。

但是，系數(shù)是正值而邊際效應(yīng)是負(fù)值，這可能嗎？

我設(shè)法制作了一個數(shù)據(jù)集，得到了以下結(jié)果。

1. reg => positive
2. ivreg => positive
3. probit coefficient => positive
4. probit marginal effect => positive
5. ivprobit coefficient => positive
6. ivprobit marginal effect =>?negative

1. 回歸 => 正

2. ivreg => 正

3. probit 系數(shù) => 正?

4. probit 邊際效應(yīng) => 正?

5. ivprobit 系數(shù) => 正?

6. ivprobit 邊際效應(yīng) => 負(fù)

This is strange in so many levels. #5 and #6 having different sign is strange. #2 and #6 having different sign is strange.
And if this is possible, in which situation would it occur?
And if this is possible, shouldn't there be somewhere in the range of x such that ivprobit marginal effect is positive?
And how should I interpret this? Is X causing increase in Y or decrease in Y? All others are positive & significant. Only marginal effects from ivprobit is negative (and sometimes significant)

這在很多方面都很奇怪。#5 號和 6 號的符號不同很奇怪。#2 和 #6 的符號不同也很奇怪。

如果有可能，在什么情況下會出現(xiàn)這種情況？

如果有可能，難道不應(yīng)該在 x 范圍內(nèi)的某個地方出現(xiàn) ivprobit 邊際效應(yīng)為正的情況嗎？

我該如何解釋？X 是導(dǎo)致 Y 增加還是 Y 減少？所有其他變量都是正的、顯著的。只有 ivprobit 的邊際效應(yīng)是負(fù)的（有時是顯著的）。

Clyde Schechter：（回復(fù)Edmondo）

I'm unable to respond to your specific question because I do not use instrumental variables in my work and I have only a very limited understanding of how they work.
I can say that with linear models, the marginal effect is equal to the coefficient. With non-linear models, the marginal effect must be conditioned on particular values of the predictor variables (or averaged over the distribution of the predictor variables) and can vary considerably, whereas regression coefficients are unconditional. So there is no necessary type of agreement between a regression coefficient and all of the infinitely many marginal effects associated with that variable.

我無法回答您的具體問題，因為我的工作中不使用工具變量，而且我對工具變量的工作原理了解非常有限。

我可以說，在線性模型中，邊際效應(yīng)等于系數(shù)。對于非線性模型，邊際效應(yīng)必須以預(yù)測變量的特定值（或預(yù)測變量分布的平均值）為條件，而且可能會有很大的變化，而回歸系數(shù)是無條件的。因此，回歸系數(shù)與與該變量相關(guān)的無限多的邊際效應(yīng)之間并不存在必然的一致。

Teresa Schützeichel：

Following this thread, I am still confused as to why the significance levels of the ivprobit original coefficients and those of its average marginal effects should not be identical. I ran an ivprobit which returned positive and significant coefficients (varying p-values for the different specifications), however the corresponding average marginal effects after running margins, dydx(*) predict (pr) were all positive and insignificant (p>0.1 for all specifications).
The other thing is the command 'margins, dydx (*) predict (pr)' returns the average marginal effects of the instrumental variable as well, ideally this should not happen..any idea why this is the case?
Thanks in advance.

跟隨著這個主題，我仍然感到困惑的是，為什么 ivprobit 原始系數(shù)的顯著性水平與其平均邊際效應(yīng)的顯著性水平不一致。我運(yùn)行了 ivprobit，得到的系數(shù)為正且顯著（不同規(guī)格的 p 值不同），但運(yùn)行 margins, dydx(*) predict (pr) 后得到的相應(yīng)平均邊際效應(yīng)均為正且不顯著（所有規(guī)格的 p 均大于 0.1）。

另外，"margins, dydx (*) predict (pr) "命令也會返回工具變量的平均邊際效應(yīng)，理想情況下不應(yīng)該出現(xiàn)這種情況。

在此先表示感謝。

Richard Williams：（回復(fù)Teresa）

Teresa, welcome to Statalist.

i suggest reading the Statalist FAQ, esp. point 12 about asking Qs effectively. Showing exactly what you typed and how Stata responded can make it easier to see what you are talking about.

Also, while I might add a link to old threads, I like to start a new thread rather than add on to a long older one. If something has 20+ posts I usually don’t want to go to the trouble of getting up to date on what has already been talked about.

i personally do not get too surprised about difference in significance levels between coefficients and marginal effects. Marginal effects can be computed in many ways, e.g. atmeans, asobserved, or at values chosen by the user. These different ways might produce different significance levels. I usually just focus on the significance of the coefficients.

特雷莎，歡迎來到 Statalist。

我建議您閱讀 Statalist 常見問題解答，尤其是關(guān)于有效提問的第 12 點(diǎn)。準(zhǔn)確顯示您輸入的內(nèi)容以及 Stata 是如何回應(yīng)的，可以讓人更容易理解您在說什么。

另外，雖然我可能會添加舊主題的鏈接，但我喜歡啟動一個新主題，而不是添加到一個很長的舊主題中。如果一個主題有 20 多個帖子，我通常就不想再費(fèi)力去了解已經(jīng)討論過的內(nèi)容。

我個人對系數(shù)和邊際效應(yīng)之間顯著性水平的差異并不感到太驚訝。邊際效應(yīng)的計算方法有很多種，如按平均值、按觀察值或按用戶選擇的值。這些不同的方法可能會產(chǎn)生不同的顯著性水平。我通常只關(guān)注系數(shù)的顯著性。（Williams最后的結(jié)論）