Questions tagged [errors-in-variables]

Errors in variables are measurement errors which increase the estimation variance (error in the dependent variable) or bias the regression coefficients towards zero (error in the independent variables).

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

Linear Regression but the Variables have errors

I have received this confusing task: You have two variables 𝑥 and 𝑦, where y is a response variable which can be written as an explicit linear function of 𝑥. However, the technique used for ...
1 vote
0 answers
10 views

What does Deming regression estimate?

Least squares regression estimates conditional means. Least absolute regression estimates conditional medians. Quantile regression’s estimate conditional quantiles. Analogously, what does Deming ...
  • 39.9k
0 votes
0 answers
24 views

Why does the IV estimator fail to attenuate the EIV bias in a two-stage FM regression?

In Jegadeesh et al. (2019), they proposed to use the instrumental variables (IVs) estimation approach to attenuate the errors-in-variables (EIV) bias, which is inherent to a two-stage Fama-MacBeth (FM,...
  • 1
0 votes
0 answers
17 views

Uniform measurement error in "errors in predictors" regression

I'm working with cancer incidence data that uses a range of ages (e.g. <1, 1-4, 5-10, ...) rather than a single value. I want to fit a model where age is a predictor. As a result, I'm curious ...
0 votes
0 answers
35 views

Measure of goodness-of-fit in errors-in-variable regression

I have two observed time series $x_i$ and $y_i$ and I want to test if $x_i$ is a good predictor of of $y_i$. So I would usually run a simple linear regression Y ~ X and use $R^2$ as a measure of ...
  • 1
1 vote
1 answer
55 views

How to test for correlated errors in regression

I understand that one assumption that must hold for regression is for there to be no correlation in the error structure. Put another way: The residuals should be impossible to predict above chance. ...
0 votes
0 answers
51 views

How to compare distributions with errors on the data points?

Here's a mock set-up of my problem: I have two non-normal probability density distributions (PDFs), $A$ and $B$. Distribution $A$ has error measurements for each data point while distribution $B$ ...
1 vote
1 answer
213 views

MLE on Structural VAR

I have a simple model that I wish to fit using data. The model is of the form below. \begin{gather} y_t = -\lambda r_t + \theta a_t + \varepsilon_1 \\ \\ \pi_t = \pi_{t-1} + w y_t + \varepsilon_2 \\ \\...
  • 117
0 votes
0 answers
162 views

Nonlinear goodness of fit for data with uncertainties in both x and y

I have a set of samples with an age given with some uncertainty (x $\pm$ dx) and that have some measured property with some given uncertainty (y $\pm$ dy). I would like to fit a curve that begins at $(...
  • 49
1 vote
1 answer
61 views

Removing the bias from some unknown measurement error

Imagine I have two variables X and Y which have a statistical relationship. However I cannot observe X. I can only observe X* = X + U where U is some 0-centered random noise. I don't know U but I ...
3 votes
1 answer
403 views

Linear least-square fitting of two variables with uncertainty on both

I am trying to find an R function to calculate the linear least-square fitting of two variables when both have an error (expressed as standard deviation). I have found this problem referred to in half ...
  • 91
0 votes
0 answers
38 views

Interview question (incomplete): extension of linear regression (errors in variable)

Here is a interview question I head from others, but I think the information may be not complete and correct. Could anyone help me to modify it? Question: Suppose $X\sim N(0,1), \epsilon\sim N(0,1)$ ...
  • 1,129
2 votes
2 answers
143 views

If the $\varepsilon$ in $Y = \beta_0 + \beta_1 X + \varepsilon$ does not represent measurement error in $Y$, then what does it represent?

The classical simple linear regressoion model is $$ Y = \beta_0 + \beta_1 X + \varepsilon. \tag{1} $$ On page 3 of these slides, the author says if there are measurement errors in the outcome then we ...
7 votes
1 answer
199 views

In reality, there is almost always measurement error in the independent variable(s), so why is this ignored in almost every linear regression model?

In the vast majority of cases, linear regression models are used in practice as opposed to the more complicated errors-in-variables models. For the sake of example, consider modelling height $Y$ vs ...
  • 735
2 votes
0 answers
52 views

Regression problem with "error in variables"

Suppose that there is a deterministic relation $y_t=ax_t$ where $x_t,y_t$ are real sequences or real functions and $a$ a constant. But only $X_t=x_t+e_t$ and $Y_t+u_t$ can be observed, with $e_t, u_t$ ...
1 vote
0 answers
74 views

How to do Error in Variables regression with known standard errors

I need some help with EiV regression and comparison of two methods. I have used two different methods to estimate the size of the same population and would like to find out how good method 1 is ...
  • 11
4 votes
2 answers
2k views

Inverse Regression vs Reverse Regression

I'm aware there's a great number of questions which deal with the mathematical difference between the two, but I'm still confused as to best practice. Basically I'm looking at a situation where we ...
  • 51
0 votes
1 answer
551 views

Big outlier in dependent variable

I have my data from the official statistics office of my country and I rechecked multiple times already. I have a big outlier skewing all my glm (poisson) modells to the extreme (like 5 times the ...
1 vote
0 answers
91 views

What kind of statistical analysis is required to compare two methods for regression

I want to do comprehensive study of errors in variables and compare the results with regression for selected parameter estimation problems in my domain where it is expected to perform better in terms ...
0 votes
0 answers
26 views

Errors-in-variables with correlated latent variable

Suppose I have data generated as follows: $\tilde{X} = k \cdot X + u$, where X is an unobserved latent variable (say the temperature of the room) and X_tilda is the observed variable (say temperature ...
3 votes
1 answer
647 views

meaning of error term being correlated with regressor

I have encountered the statement that "the error term and one of the regressors are correlated" a few times and I am having trouble understanding what is meant exactly. Let's say we have a DGP $$y=\...
  • 181
3 votes
2 answers
1k views

Including model uncertainty in non-linear least squares minimization

The problem I have experimental data $Y$ with heteroscedastic and normally distributed uncertainties characterized by covariance matrix $C_{exp}$. I want to fit the data using model $F(X, \beta)$ ...
  • 49
1 vote
1 answer
129 views

Difference between estimating parameters for prediction and estimating parameters for their own sake

In a 1989 paper on orthogonal regression, Ammann and Van Ness write: An important caveat should be noted. The errors-variables-model is useful when the primary goal is to estimate the model ...
1 vote
1 answer
406 views

Statistical library for orthogonal distance regression with a ridge penalty?

There are many libraries in R and python for doing orthogonal distance regression and for doing ridge regression separately. Is there one for doing them at the same time?
1 vote
0 answers
140 views

Correct error estimation for linear fit

This may be a simple problem, but I want to be thorough in setting up my problem as I'd like to know why I should proceed in one of two ways (or another if someone thinks it is suitable), so please ...
2 votes
0 answers
17 views

Exponential errors in variables model with known uncertainties

I have $N$ data points that I am trying to fit using a function of the form $y_i = \prod_j {X_{i,j}}^{b_j}, \quad j=1..N$ where $\mathbf X$ and $\mathbf y$ are measured values. The form of this ...
1 vote
0 answers
28 views

How would you attempt to distinguish measurement error from a true distribution but no "true" singular value?

I am self-taught amateur in statistics and I have confusion is based calculations with error-in variables with the added object being measured having a no "true" value but a "true" range. For me, ...
6 votes
1 answer
313 views

If in this problem I regress $x$ on $y$ instead than $y$ on $x$, do I need to use an error-in-variables model?

I was trying to write an answer for this question: Selection of data range changes coefficients too much in lmer (inverse regression) Basically the OP has lots of data of Amplification vs Voltage (...
  • 16.5k
3 votes
2 answers
319 views

Find error interval of linear relationship

I have two sensors of different quality capturing the same process, where one of them is much more accurate than the other. Hence, I want to find out how much better. Let us for example say that the ...
1 vote
0 answers
59 views

Fitting a logarithmic growth curve with error (an interval) of the explanatory variable?

I have a series of human growth data that I wish to fit to a 3 parameter logarithmic growth curve: s(i) = Beta0 + B1*T + B2*ln(t), where s is a length and t is an age. The only problem is that this ...
2 votes
0 answers
116 views

Rotation changes correlation - correction from OLS

Let $X,Y$ be real random variables with finite variances, and with no loss of generality assume $\mathbf{E}[X] = \mathbf{E}[Y]=0$. For simplicity, I will focus on the case $\mathrm{Var}X \neq \mathrm{...
9 votes
1 answer
2k views

Control Function Approach and Bootstrap

Let's start assuming that I have cross-sectional data on $y$, $x_1$, $x_2$ (see below for $y$, $x_1$, $x_2$). I want to estimate the effect of variables $x_1$ and $x_2$ and their interaction ($x_3= ...
0 votes
0 answers
612 views

What to do when expectation maximization results are invalid (they don't match the likert scale)

I have missing values (MCAR) for which I used EM to fill in those values. Some of the imputed values are negative integers or zero. I am using a likert scale to measure responses, and thus i need the ...
3 votes
0 answers
96 views

small sample approach to simple linear regression with errors-in-variables (measurement errors)

I seek to estimate $b_1$ and $b_0$ from data of the form: $$y_i = b_1x_i + b_0 + e_i, \quad i\in\{0,1,...,N-1\}$$ given $\{y_i\}$ and $\{\tilde{x}_i\}$ where $\tilde{x}_i=x_i + n_i$ (i.e., error-in-...
  • 205
1 vote
1 answer
36 views

correlation when x and y are uncertain

Suppose that for $1\le i \le N$ $$\begin{align} Y_i^j &= f(X_i) + \epsilon_y \qquad &1 \le j \le R_y^i \\ Z_i^j &= af(X_i)+b + \epsilon_z \qquad &1 \le j \le R_z^i \end{align}$$ where $...
  • 842
2 votes
0 answers
61 views

How do I use American Community Survey income estimates and standard errors in an error-in-variables model?

Income distributions tend to be highly skewed. Yet the American Community Survey (ACS) provides only two summary statistics for its median income estimates in small areas: the mean estimate and a ...
5 votes
1 answer
2k views

How do errors in variables affect the R2?

I've got a question about errors in variables. So, if I run a standard linear regression to estimate b in y = a + bx, but my ...
  • 225
2 votes
0 answers
29 views

Dealing with independent variables where each point is an coefficient

I want to test if a behavior is influenced by population size. The former is measured as a continuous normal variable. The latter is estimated using Schnabel's method, and thus each sample has a ...
  • 447
0 votes
1 answer
203 views

OLS Assumptions - Errors are normally distributed

I am currently working on a research project based on the data of a big survey. I derived a variable set, which I would like to investigate. Before starting with it, I would like to check the ...
3 votes
0 answers
81 views

Multiple errors-in-variables regression with collinearities

I have a $[k \times N]$ matrix of predictors / independent variables and a $[k \times N]$ matrix of predictands / dependent variables. I have uncertainty estimates for each predictor and each ...
  • 1,407
5 votes
0 answers
132 views

Intuition: Why do I have to worry about errors-in-variables?

I've read that (ordinary) linear regression assumes that there are measurement errors in the dependent variable, but no measurement error in the independent variables -- and if I have measurement ...
  • 6,263
2 votes
0 answers
151 views

Standard error of mean - under measurement error

I have two observations of a normally distributed random variable: $$ X_1 = 0.02 \\ X_2 = 0.10 $$ Obviously the sample mean equals 0.06, and the standard error of the mean (SEM) is equal to 0.04. ...
2 votes
1 answer
2k views

Bayesian errors-in-variables model definition in JAGS and symbolically

I'm fairly new to probability theory and am attempting to understand and implement an errors-in-variables simple linear regression model. I am assuming a model of the form $$ Y=\theta X_a+\epsilon_Y ...
  • 85
3 votes
0 answers
1k views

Model selection: OLS vs TLS

I have two sets of real-valued data and I am interested in their correlation. From my perspective, there appear to be errors both variables, so I am inclined to perform a regression with TLS (Total ...
0 votes
1 answer
272 views

Fit error-in-variables polynomial regression using mle2 (R)

I need to fit a polynomial regression that accounts for measurement errors. I found out how to do it with a mcmc model (using RJags) and I would like to do it with a Maximum Likelihood Estimator (...
  • 131
3 votes
1 answer
1k views

Total least squares with weights [duplicate]

I am looking for a way to perform weighted total least squares in R. I know one can use PCA for this as described nicely in the following post. How to perform orthogonal regression (total least ...
  • 321
0 votes
0 answers
226 views

linear regression accounting for standard errors of variables

I am trying to figure out what is the best way to estimate beta why accounting for the uncertainty in x and y. For example, I have ...
  • 321
2 votes
0 answers
39 views

Signing the inconsistency of a control functions estimator with an incomplete set of instruments

I've got a model of the form $$ y = 1\left(\alpha+\delta x+\beta_1\tilde v+\tilde\epsilon>0\right) $$ $x$ is endogenous, and $\tilde v$ is a control function residual: $$ \tilde v = x-Z'^{inc}\...
  • 12.6k
8 votes
4 answers
3k views

Systematic/measurement error on a linear regression

Suppose I have a set of data ${(x_i,y_i)}$ in which the uncertainty in the measurements ${(\Delta x_i,\Delta y_i)}$ (which come from the propagation of systematic errors from the measurement apparatus)...
5 votes
0 answers
178 views

Total least square intuition

I have yet to find a good intuitive explanation of TLS. Online resources tend to focus on the vertical vs. perpendicular square error pictures (I don't need to see perpendicular lines to understand ...