Questions tagged [total-least-squares]

A technique to estimate parameters $\beta$ of the linear model $Y=X\beta$ when both $Y$ and $X$ are subject to measurement error. Includes Orthogonal and Deming regression as special cases.

Filter by
Sorted by
Tagged with
0 votes
0 answers
39 views

Orthogonal regression that minimizes absolute error instead of squared error

Fitting a line through 3D points is usually done by orthogonal least squares (aka "total least squares"), i.e., by minimizing the mean squared orthogonal distance between the points and the ...
user avatar
  • 3,021
0 votes
0 answers
18 views

How to do weighted Total Least Squares (TLS, aka Orthogonal) Regression faster in R?

I am working on a problem where I am using orthogonal regression (aka Total Least Squares (TLS)). I originally used the odregress() function in R from the pracma package. However, it doesn't allow ...
user avatar
1 vote
0 answers
9 views

duel regression with orthogonal constraints on coefficients [closed]

I'm trying to solve a problem where I need to find two deming style regression models onto two different data sets of equal dimension who's coefficients(A,B) satisfy the following criteria with ...
user avatar
  • 11
0 votes
0 answers
15 views

Total least square and jackknife method

I am trying to understand the total least square. I have found in example here, unfortunately it is only in German, but the most important thing are the five example data points. These are (1,3), (3,1)...
user avatar
  • 3
0 votes
0 answers
85 views

Interaction effects in Deming/orthogonal/total least squares regression

I am interested in analysing the difference in slopes between two continuous variables (x & y) in 2 groups, pre and post intervention. ...
user avatar
1 vote
0 answers
29 views

Propagating uncertainty in least squares regression

I am currently running a nonlinear regression on $y=f(x,\theta)$, the variable $x$ being a known input and $y$ being a measurement result. I would like to estimate the parameters $\theta$ as well as ...
user avatar
1 vote
0 answers
23 views

How to handle nondifferentiable points of the objective function in the geometric circle fit?

We a given measured points in $(x_i,y_i) \in \mathbb{R}^2$, $i=1,..,n$. The geometric circle fit is the circle with center (a,b) and radius $r$ that minimizes the squared Euclidean distances between ...
user avatar
0 votes
1 answer
98 views

Errors only in variables model, and polynomial fitting

I have a bunch of data points $(x, y)$, and I know that they fit well to a model of the form $y = a + bx + c x^2$, with $a \approx 0.01, \ b \approx 1\ \textrm{and}\ c \lesssim 0.1$. I'd like to fit ...
user avatar
  • 101
2 votes
1 answer
118 views

Use of ordinary least squares line in correlation analyses

I want to study and plot the correlation between two variables, X,Y. Both are measured, so they have comparable noise. Correlation (and not regression) is the correct analysis here. In a paper, I ...
user avatar
  • 388
1 vote
0 answers
69 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 ...
user avatar
  • 11
2 votes
1 answer
44 views

Linear regression with normally distributed data and model with arbitrary covariance

Consider the linear regression problem $$(A+\Delta A)x = b + \Delta b.$$ If $\Delta A = 0$ and $\Delta b$ is identically and independently distributed, then ordinary least-squares gives a good (BLUE) ...
user avatar
  • 131
1 vote
1 answer
34 views

Which is more correct to calculate least squares (errors) during nonlinear curve fitting, calculating after transformation or in the original form?

I am studying curve fitting and linear regression. I am supposed to find a and b in the equation $$ P=ae^{bh} $$ so I transformed it to $$ lnP=ln a +bh $$ then $$ Y=c+bX $$ after that I solved it to ...
user avatar
  • 13
0 votes
0 answers
52 views

Identifying an algorithm described as 'Tukey approach' for ignoring outliers?

MVTec's machine vision library Halcon has an operator fit_line_contour_xld for robustly fitting lines to 2D points. Here's the documentation entry for that operator:...
user avatar
3 votes
2 answers
893 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 $F(X, \beta)$ ...
user avatar
  • 49
0 votes
0 answers
24 views

Standardizing regression coefficients in the case of Total Least Squares Regression

So I know that there exits a method to standardize regression coefficients by simply multiplying them with the ratio of the standard deviations of the two variables. I'm wondering if this holds true ...
user avatar
1 vote
1 answer
83 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 ...
user avatar
1 vote
1 answer
334 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?
user avatar
1 vote
0 answers
1k views

Difference between "orthogonal distance regression" and "total least squares"

I'm trying to figure out the difference(s) between total least squares (TLS) and orthogonal distance regression (ODR). Both techniques are used when there is error in the dependent variable. Per this ...
user avatar
1 vote
0 answers
42 views

least absolute deviation version of deming [closed]

Based on what I know, deming minimizes the sum of square of perpendicular distance to the regression line. Is there a package in R that can run regression that minimize the sum of absolute value of ...
user avatar
  • 111
3 votes
2 answers
1k views

Why my Deming Regression line change so much when switching variables? If they seem to be a linear relationship betwen them?

I am trying to fit a line that best predicts the production of energy Y given the speed of wind X, a typical Y = xm + b , using deming regression. I am looking for the slope and the intercept of that ...
user avatar
1 vote
1 answer
40 views

Correct way to fit a line in 3D (x-position vs y-position vs other quantity)

I have measured the position of light spots $(x,y)$ in an arbitrarily chosen basis and I compare that to some other measured quantity, say the brightness of each spot $B$. Now, in theory all the ...
user avatar
  • 123
-1 votes
1 answer
415 views

Deming regression prediction interval using jackknife resampling

I am trying to write a custom Deming function following the maths in Linnet (1993): https://www.ncbi.nlm.nih.gov/pubmed/2281234 Using jackknife resampling I calculate the standard error for the ...
user avatar
  • 115
1 vote
1 answer
504 views

Standard error of coefficient estimates for model II regression

I'm working with time series data that has error in both the dependent and independent variables, so I'm analyzing each half hour of data with model II linear regression, specifically geometric mean ...
user avatar
0 votes
1 answer
289 views

Selecting appropriate likelihood during non-linear regression

When performing regression to fit a function, $f (x,{\bf \beta})$, to a set of observed data, $y_i(x_i)$, we are seeking to optimize the parameters, $\beta$, of the fitting function, to minimize some ...
user avatar
  • 447
1 vote
0 answers
33 views

Can total least squares be used to account for uncertainty due to measurement timing?

Suppose we have a dataset in which we wish to perform regression analysis, and where the response/dependent variable is a measurement at time T. But due to pragmatic sampling we do not have ...
user avatar
  • 413
0 votes
1 answer
56 views

Regression to estimate parameters

I would like some suggestions to tackle the following problem. Given a system $y = X\beta$ where $y \in \mathcal{R}^m$, $X \in \mathcal{R}^{m \times n}$, $m\geq n$, and $\beta \in \mathcal{R}^n$, ...
user avatar
  • 131
2 votes
1 answer
235 views

Orthogonal polynomials + cross validation: should subsetting be done prior or after constructing the orthogonal polynomials?

So, just to start... I've just learned of orthogonal polynomial regression today. I've gone through the master's-level linear models courses, and we did not cover that topic. I was always under the ...
user avatar
  • 3,883
4 votes
2 answers
1k views

Why does the total least squares line in 2D pass through the average across all data points?

I have $N$ data points $\mathbf{m_k}$ and I want to fit a line through them with minimal error $$J = \sum_k^N ||\mathbf{m_k^*} - \mathbf{m_k}||^2 = \sum_k^N ||\mathbf{m_0} + a_k\mathbf{e} - \mathbf{...
user avatar
  • 1,226
1 vote
0 answers
797 views

Advantage of orthogonal polynomials

What is the sense or background of orthogonal polynomials (regarding using mixed models)? I would like to know why they shall or should be orthogonal. Is it to build independent sample points? On Is ...
user avatar
  • 2,197
2 votes
1 answer
597 views

Least Squares Derivation

I come from physics and would like to derive the chi-square function given by the Particle Data Group: \begin{equation} \chi^2 (\boldsymbol\theta) = (\boldsymbol y-\boldsymbol\mu(\boldsymbol \...
user avatar
  • 131
2 votes
0 answers
153 views

Deming regression implementation: force intercept to 0

I implemented Deming Regression in a known programming language, using the algorithm from here: https://en.wikipedia.org/wiki/Deming_regression However, the algo does not specify what to do in case we ...
user avatar
  • 151
4 votes
0 answers
215 views

What's the shape of confidence interval for linear regression estimated through TLS (total least squares)?

So, we know the shape of confidence intervals for vanilla linear regression estimated through OLS (ordinary least squares): Shape of confidence interval for predicted values in linear regression. In ...
user avatar
  • 15.7k
10 votes
1 answer
1k views

When to use Deming regression

I am currently working on a way to transform two different phosphorus test values into each other. Background There exist many (extraction) methods to measure plant available phosphorus in soil. ...
user avatar
  • 101
3 votes
1 answer
6k views

Generating Random Data Sets for Linear Regression with Random Slope and Error Term in R

I want to test the effects of sample size on Deming regression using simulated paired data in R. As the data are paired, the expected slope value should be 1 and the intercept 0. The code I have ...
user avatar
  • 115
2 votes
0 answers
178 views

Specifying the Error Ratio in Deming Regression

We use Deming Regression for method validation in our clinical ELISA laboratory. As we run clinical trials this needs to be performed if we get a new reagent batch, batch of plates, new plate reader ...
user avatar
  • 115
2 votes
0 answers
516 views

How to compare two Deming regressions?

I have two small data sets. For each data set, I performed a Deming regression. Hence, I have, for each data set, the relationship between X and Y in the form of slope+intercept coefficients. Now, I ...
user avatar
3 votes
0 answers
94 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-...
user avatar
  • 205
1 vote
0 answers
50 views

Regression of dataset with many x values per y value

I am performing a total least squares regression in which I have many x observations for a given y observation. The x observations are normally distributed. I am aware I could do some sort of ...
user avatar
  • 147
0 votes
1 answer
454 views

why regress a y variable on a vector of just ones

I am trying to understand some code somebody has written. I have a y vector which is a factor say book to price for 500 companies. Then I also have a x vector of the same shape which is just ones. ...
user avatar
  • 123
4 votes
2 answers
461 views

Find best fit diagonal matrix for error minimization

I want a set of input values to be as similar to the output values as possible. I have an input matrix X (m*n) that has m data points and n dimensions for each data point. I also have an output matrix ...
user avatar
  • 63
0 votes
0 answers
654 views

What weights for weighted Total Least Square (TLS) regression?

I have a dataset with known errors in both the X and Y and want to perform a simple linear regression. From reading other posts, it seems I want to TLS over OLS due to the presence of error in both ...
user avatar
  • 147
3 votes
0 answers
80 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 ...
user avatar
  • 1,397
2 votes
0 answers
51 views

$X$ and $Y$ at different scales in total least squares?

As this section of the Wiki article says, the best (in total least squares sense) matrix $B$ that projects $X$ to $Y$ is given by $$B=-V_{XY}V_{YY}^{-1},$$ where $V_{XY}$ and $V_{YY}$ come from the ...
user avatar
2 votes
1 answer
133 views

Is quantile regression a better option than total least squares RMA in this case

I have paired cobalt concentrations in bird blood and feathers. Blood levels give me an idea of how recent the cobalt exposure was (<30days), feather give the 6 month accumulated total. Previous ...
user avatar
2 votes
0 answers
518 views

Confidence/prediction intervals for total least squares regression

I am learning the ropes of total least squares regression and I found this thread How to perform orthogonal regression (total least squares) via PCA? where the answer by @amoeba, together with some R ...
user avatar
  • 161
16 votes
2 answers
9k views

Regression when each point has its own uncertainty in both $x$ and $y$

I made $n$ measurements of two variables $x$ and $y$. They both have known uncertainties $\sigma_x$ and $\sigma_y$ associated with them. I want to find the relation between $x$ and $y$. How can I do ...
user avatar
1 vote
0 answers
96 views

Introductory references about total least squares

I am an engineering student and I've been recently told about the Total Least Square (TLS) method. I am interested in applying it to topographic measurements and in comparing the results with the ...
user avatar
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 ...
user avatar