Questions tagged [identification]

The tag has no usage guidance.

Filter by
Sorted by
Tagged with
1
vote
2answers
33 views

Simultaneity in causal diagrams

Lets assume we have simultaneity problem. Variable x causes y and y causes x. As an example i would state alcoholism: the more respondent consumes alcohol, the more 'is' alcoholic (measured for ...
1
vote
1answer
70 views

Can't compute standard errors: the information matrix could not be inverted

I am trying to compute a Structural equation model. I have identified one model, in which I have included all the subscales simultaneously. However, when I try to look at the subscales separately, I ...
1
vote
0answers
17 views

How to verify identification of a model?

I have the system of simultaneous equations: $$ \begin{cases} y_1 = b_{12}y_2 + b_{13}y_3 + a_{11}x_{12} + a_{13}x_3 \\ y_2 = b_{21}y_1 + a_{21}x_1 + a_{22}x_{2} \\ y_3 = b_{32}y_2 + a_{31}x_1 + a_{...
1
vote
0answers
14 views

What is an intuitive of definition of “point identification” (point identified parameter) in econometrics?

I've recently come across the notion of point identification in several econometric papers. See, e.g., https://scholar.harvard.edu/files/tamer/files/pie.pdf, who mentions point identification ...
2
votes
1answer
43 views

Can anyone show how the concept of Identifiability is geometrically/intuitively presented?

The motivation for this question comes from the following: When I was studying statistics for the first time long ago, no one presented the mathematical concepts behind linear regression, like the one ...
0
votes
0answers
10 views

Instrumental variables, do we only need cov(z,e)=0 for exogeneity, and not E[e|z]=0?

Looking at the formula for instrumental variables estimator, we have : B_iv = B + (z'x)^-1(z'e) and then taking probability limits it becomes evident that we need cov(e,z)=0 and cov(z,x)=/=0.. but ...
0
votes
0answers
14 views

What kinds of variation am I able to explore using panel data with 3 fixed effects?

this question is similar to an unanswered question of mine. My data explores variation at three possible dimensions: $a$ area, $s$ sector, and $t$ time. There are 400 areas, 13 sectors and 8 time ...
4
votes
3answers
92 views

Parameter identification v. causal identification

As others, it seems (Identification of parameters problem), I get confused about the use of the word "identification" in econometrics. It seems some people talk about "identification" in the sense ...
0
votes
0answers
23 views

Classification - deterministic to probabilistic

Let $\mathcal{M} = \{X_1 , ... , X_N\}$ be a collection of objects, and assume that $x = X_i$. Imagine that we cannot observe $x$ directly, but we do have measurements $y = y(x)$ (only 1 dataset, ...
0
votes
0answers
15 views

Identify Effects in an Age-Cohort-Period Model

I have a survey dataset that contains the risk attitudes on individual levels for three years (2004,2009,2014). The research goal is to check the consistency of those attitudes over time using a ...
1
vote
0answers
13 views

Can I combine equations to produce overidentifying restrictions?

Say theory tells me that $$ y = f(x_1,x_2|\theta_f) $$ where $\theta_f$ is a set of parameters Similarly, theory tells me that $$ y = g(x_1,x_3|\theta_g) $$ where $\theta_g$ is another set of ...
3
votes
2answers
107 views

Identification from implicit function

Suppose my observed data $y$ and $x$ is generated by the following relationship for each observation $i$: $$ y_i = h(y_i,\theta) + x_i + \varepsilon_i$$ where $x_i$ is a strictly exogenous variable ,...
1
vote
1answer
83 views

Inconsistent Ljung-Box test result and plot of autocorrelation function of residuals

I get an inconsistent result for the Ljung-Box test: in fact when I run it using the Box.test function it doesn't make me reject the null hypothesis of residuals being white noise, but when I plot the ...
2
votes
0answers
174 views

Fitting a Beta distribution only using coin flips from the biased coins it generates

I have a Beta distribution $D$ with unknown parameters $\alpha$ and $\beta$ which I wish to estimate. If I was given samples $p_1, \ldots, p_n$ from $D$, then it's relatively straightforward to fit $...
1
vote
1answer
95 views

Identification of discrete choice models

Consider the classical Logit model. In particular, let $\mathcal{Y}\equiv (0,1,...,L)$ be the set of options available to consumers, where $0$ denotes the outside option. Let $$ u_y\equiv \begin{...
1
vote
1answer
290 views

Do I need Sargan test with equal numbers of instruments and endogenous variables when one instrument can affect more than one endogenous variables?

I have a instrumental variable logistic regression I run with three instruments (z1, z2, z3) and three endogenous variables (k1, k2, k3). Therefore, since the number of instruments "3" equals the ...
0
votes
1answer
57 views

Verifying Identification Results for Univariate Regression

So I have this linear regression model shown below and I'm supposed to be showing that equation 3 is equal to equation 4. There's a hint that says a 2x2 inverse matrix appears in the proof, but the ...
1
vote
0answers
94 views

Overidentified vs just identified models

Why go with overidentified models as opposed to just-identified? If you can go with over-identified models, how many instrumental variables can you have at max in a 2SLS model?
1
vote
0answers
2k views

Stata: Difference in Difference; common-trend plot [closed]

I estimate the following flexible difference-in-difference model with Stata: regress Y i.time i.time##i.treatment Starting from this model, I'd like to ...
1
vote
1answer
56 views

What does each cluster represent?

I have a dataset from a questionnaire with over 10000 rows and 30 variables. I am trying to have an insight of the data so I tried to cluster similar items. I first made a dimension reduction ...
2
votes
0answers
42 views

Why is uncorrelated(exogeneity) “good enough” for identification in regression?

Let the model $y=x+x^2-1$ be exactly correct, where $x\sim N(0,1)$, then $x^2\sim \chi^2_{(1)}$. Say we want to estimate the model $y=\beta x-1+\epsilon$ by least squares. Let $(y_i,x_i)_{i=1}^n\...
0
votes
0answers
31 views

Identification and estimation of my structural model with a latent variable

I am having some trouble trying to identify the parameters in the following structural model that I am trying to estimate. $$ y = a'x_1 + \beta\eta + \epsilon_1 $$ $$ \eta = b'x_2 + \delta T+\...
1
vote
0answers
30 views

Alternatives to calculating the rank of the information matrix in determining if the model is identifiable

I have a known non-linear model $h \in \mathbf{R}^n$: $$ y = h(\theta) + \epsilon, $$ where $\theta\in \mathbf{R}^m$ is a parameter vector, and $\epsilon$ is a normal random variable with zero mean ...
3
votes
0answers
73 views

Weak instrument problem, how can a “false experiment” serve as an identification check?

In their 2004 paper, Miguel et al. investigate the role of income growth by using current and lagged rainfall as an instrument. However, the instruments are somewhat weak: the joint F-statistic is ...
1
vote
0answers
21 views

What exactly does “local identification” mean in the context of structural VARs and VECMs?

In Lutkepohl's "New Introduction to Time Series Analysis", chapter 9, this term is frequently mentioned, but is in no where well-defined. So what exactly does "local identification" mean in this ...
0
votes
0answers
306 views

Create Gaussian noise for artificial dataset with different noise levels

I am creating an artificial dataset corresponding to different noise levels. This is to simulate results of a recognition software (e.g. face recognition). For example, for $noise_{level} = 0.1$, the ...
0
votes
0answers
51 views

Full Support of r.v. and model identification

Consider the model: $$y_i=x_i'\beta + \epsilon_i$$ where $\epsilon_i$ is i.i.d. and follows a $N(\mu,\sigma^2)$ with $\sigma^2>0$ and where the distribution of $x_i$ has full support on $R^k$. You ...
1
vote
1answer
592 views

Confirmatory factor analysis model identification

The minimum number of indicators for a single factor measurement model in CFA is 3. This follows from k(k+1)/2 where ...
0
votes
1answer
1k views

What do the assumptions for 2SLS (two-stage least squares) mean in the context of instrumental variables?

The assumptions for 2SLS are (z are the vector for instrumental variables; x are the explanatory variables in the model; u are the vector for the error term): Assumption 2SLS.1: E(z'u)=0, Assumption ...
3
votes
1answer
200 views

Which parameter is identified by $E(Y|X)=X\beta$?

I have been confused by the concept of parameter identification for a while now. I am asking this question mainly to test my understanding. Here is the linear regression model: $$y=X\beta + \...
3
votes
1answer
816 views

identification SEM model

Can someone explain to me how I calculate the non-redundant parameters? I read everywhere that you calculate this by p (p + 1) / 2, but do you have to do this per latent variable or in total or with ...
0
votes
1answer
33 views

How to calculate accuracy of identification using embedings and corresponded labels? [closed]

I'm trying to implement deep speaker embeding system and after getting voice embeddings I need to somehow calculate accuracy. But there is no mention in deep speaker paper about how they calculated ...
1
vote
1answer
974 views

Identify parameters for ARIMA model

I am trying to build ARIMA model, I have 144 terms in my standardized time series, which represent residuals form original time series. This residuals, on which I would like to build ARIMA model, are ...
0
votes
1answer
23 views

Why is parameter identification defined on the distribution of observables, rather than just the sample?

I'm struggling to intuitively understand parameter estimation. Specifically, why do we say that a parameter is identified if it is determined by the probability distributions of observables? e.g. we ...
0
votes
1answer
55 views

Tuning an approximate model with data?

Given an approximate model (obtained by theoretical simplifications etc.), how can observations (data) be used to fine tune it?. Standard supervised learning techniques can used for constructing ...
1
vote
1answer
1k views

ARMA vs AR process differences

According to wold representation theorem every covariance-stationary time series can be written as a linear combination of lagged values of a white noise process (MA(∞) representation). Now if a MA(∞) ...
3
votes
1answer
59 views

Identifying restriction in MLE?

Consider the following density function: $$ f(x; a,b) = \frac{(1-a)b}{2a x^2}$$ where $a$ and $b$ cannot be separately identified. Now suppose that given my data, I can restrict $a$ to two possible ...
0
votes
1answer
104 views

What is next after a referee rejects an instrumental variable strategy?

I have a paper that just got rejected. It appears that my identification strategy is not sound enough. Below I describe the critique and seek some advice. My sample is elderly aged 55 and over. The ...
0
votes
1answer
60 views

Convolutional Neural Networks for identification of specific objects instead of object categories?

Is it for example feasible to hand a query image of my dog "Dave" to a CNN, and then use the CNN to find all images of "Dave" in a database of dog images? If so, how would one do this? If not, why ...
0
votes
1answer
2k views

Identification assumptions and causal relationships

I'm new to econometrics and I'm having a hard time answering if the following statement is true or false: "In regression studies, making adequate identification assumptions is sufficient for ...
1
vote
1answer
840 views

Residual ACF of AR and MA models is the same

For Autoregressive Integrated Moving Average models (ARIMA), if the residual autocorrelation function (ACF) of both autoregressive, AR(1), and moving average, MA(1), models is the same what does this ...
0
votes
0answers
451 views

Identification with multiple fixed effects

I have been thinking about this question for a long time: I wanted to ask that if we have a fixed effect regression (say data on 30 counties over 10 years). Now, say I run a regression $Y_{it} = \...
1
vote
1answer
148 views

Which VAR model should I use

I have a set of covariance stationary variables which are slightly correlated to each other (<20%). I want to model the dependencies among the variables. I found out, that there are three types of ...
1
vote
1answer
121 views

Identifiability Versus Convexity

I'm a little unclear on the definitions of "identifiable" and "convex." Consider the case where $X_1, \ldots, X_n \overset{iid}{\sim} \text{Bernoulli}(p)$. Then our likelihood function is $L(p) = p^{\...
0
votes
1answer
164 views

Least squares system identification gives wrong coefficients

I am working on system identification using least squares method. I implemented the algorithm as recommended by the original paper. This link describes what I implemented. The image below is for ...
1
vote
0answers
37 views

Finding the probability of a Nearest Neighbour miss-identification in 8 dimensions

I'm trying to ascertain the accuracy of a device used to distinguish values from different populations. Currently each device measurement contains a data point from 8 different sensors. The value ...
0
votes
0answers
48 views

Identification of a system of equation with a parameter in the exponent

I'm trying to proof identification for the following system of $N$ equations: $$ y\times(1-\sigma) = x^{(1-\sigma)} - 1 + \alpha \times ( z^{(1-\sigma)} - 1) + \beta \times ( w^{(1-\sigma)} - 1) $$ ...
0
votes
2answers
172 views

Is the equation identifiable?

Is the following equation identified. Is it possible to obtain the estimates of parameters $\alpha$ and $\beta$ in OLS estimation. $y_t=\alpha+\beta x_t +\frac{z_t}{\beta}+u_t. $ $x_t, y_t, z_t$ ...
2
votes
1answer
331 views

Dependent gaussian processes

I need a help with gaussian processes. I am implementing dependent gaussian processes as on this paper Boyle, Phillip, and Marcus Frean. "Dependent gaussian processes." Advances in neural ...
4
votes
1answer
1k views

Understanding mean independence in the regression setting

The notion of uncorrelated ($\mathbb{E}[XY]=0$) and mean independence ($\mathbb{E}[X|Y]=0$) are mentioned in different setting of regression assumptions. We know that $\mathbb{E}[X|Y]=0$ implies $\...