Skip to main content

Questions tagged [error]

The error of an estimate or prediction is its deviation from the true value, which may be unobservable (e.g., regression parameters), or observable (e.g., future realizations). Use the [error-message] tag to ask about software errors.

Filter by
Sorted by
Tagged with
3 votes
1 answer
53 views

Subtraction of Monte Carlo integrals - Catastrophic cancellation

I am attempting to estimate a quantity $Q$ which is given by the difference between two functions of Monte Carlo integrals over some set of points $\{x_i\}_{i=1}^N$, call the estimator $\hat{Q}$: $$ \...
Eweler's user avatar
  • 402
0 votes
0 answers
24 views

How can I find the margin of error of the extrema point?

I create a 4th or 5th degree fit curve to find the extrema point of distribution. However, how will I calculate the margin of error of the extrema on x values? Is there any statistical method or ...
Erkan Güler's user avatar
0 votes
1 answer
46 views

How do I calculate the error on the extrapolation of a double natural log fit? [duplicate]

I am writing software in Python that tries to fit a data set $t, y$ to the function $y = a \ln(pt) - b \ln(qt)$ and solve for the value of $y$ at $t=30$, denoted $y_0$, and its error $\sigma_{y_0}$. ...
ohshitgorillas's user avatar
0 votes
0 answers
15 views

Can't understand this expression used for quantifying error during gradient checking

I was going through Andrew Ng's course 2 in 'Deep Learning specialization' wherein he talks about gradient checking using two sided distance for approximation. My question is more about the choice of ...
Newton's in-law's user avatar
0 votes
0 answers
28 views

How to measure the error on extrapolation from a double log fit? [duplicate]

I am writing residual gas analysis mass spectrometry software in Python. One of the functions of this software is to take the raw mass spec intensity data, $y$, and timestamps $t$, and fit them to the ...
ohshitgorillas's user avatar
-1 votes
1 answer
35 views

Calculating error on a double natural log fit

I am writing residual gas analysis mass spectrometry data reduction software in Python. The evolution of gas intensity $y$ over time $t$ in the mass spec is roughly a double natural logarithmic ...
ohshitgorillas's user avatar
0 votes
0 answers
11 views

What is the error on the weighted mean?

I am combining bins in the histogram. I have some code that uses this formula to calculate the error on the weighted mean: $$\sigma = \frac{\sqrt{\sum \frac{w_{i}(w_{i}\sigma_{i}^{2}+x_{i}^{2})}{\sum ...
manylya's user avatar
  • 23
1 vote
0 answers
20 views

Testing the difference between two Root Mean Square Error values for statistical significance [duplicate]

I would like to compare the predictive power of 2 models. The models are meant to model count data and respective probabilities. I am using two metrics as means of comparison: Root Mean Square Error ...
Astral's user avatar
  • 133
2 votes
1 answer
21 views

Is Mean Square Prediction Error acceptable to use if predicted values are continuous but actual observed values are discrete?

I would like to compare the predictive power of 2 models. The models are meant to model count data, so the actual observed values are discrete. However both models are designed such that they output ...
Astral's user avatar
  • 133
0 votes
1 answer
15 views

How to determine the confidence intervals for the principal axes of a second-rank tensor?

The question in short: How does one estimate the confidence intervals for the principal axes of a second-rank symmetric tensor when the measurement errors are themselves a function of the values of ...
Armadillo's user avatar
  • 363
0 votes
0 answers
17 views

Time series : Is SARIMA(p, 0, q)(P, 0, Q) a non-stationary model?

If the data is well explained without any differencing or seasonal differencing but requires some seasonal AR and MA terms, can we say that the data is stationary? I thought SARIMA was designed to ...
kingjerry's user avatar
5 votes
1 answer
37 views

Validating binary prediction model

Suppose we have a model that predicts for binary event $e$ ($0$ or $1$) with a single output $p$ (the expected probability $e$ occurs). If we are able to compare $p$ with the true value of $e$ ($0$ or ...
shrizzy's user avatar
  • 151
2 votes
0 answers
34 views

Errors and residuals in simple exponential smoothing (state space form) in FPP textbook

I am reading Hyndman & Athanasopoulos "Forecasting: Principles and Practice" 2nd edition (FPP2). (I am aware that 3rd edition exists.) In the chapter about exponential smoothing, section ...
Richard Hardy's user avatar
2 votes
2 answers
38 views

How to measure the error between modeled and observed data?

Consider a scenario where observed data is represented in grey and modelled data in red, as below Here, the x-axis is a position, and the y-axis is an expected time, so that the slope defines, in a ...
sam wolfe's user avatar
  • 121
0 votes
0 answers
19 views

How can I provide meaningful commentary about the uncertainty associated with a population estimate drawn form individual ML predictions?

Context: Suppose a team develops a prediction model that predicts the presence of a condition for a given individual. This model is trained and externally validated before being picked up by a ...
PC9393's user avatar
  • 1
0 votes
0 answers
16 views

How do you calculate a confidence interval if you only have an error for each input?

I have made a model (a python script), which takes as its input a scalar value $C_d$, and outputs a vector $\vec{u}$. I also know the 'real' output $\vec{u}^{\text{actual}}$ , which is calculated from ...
L.C. Huirne's user avatar
4 votes
2 answers
48 views

Machine learning benchmarks: MAE, RMSE, and R-squared

I'm working on a machine learning problem, and I'm having trouble interpreting different measures of model performance. I have a single dependent variable (proportion change between two treatments, ...
S. Robinson's user avatar
0 votes
1 answer
61 views

The error-rate in "The elements of statistical learning"

This picture is from the book "the elements of statistical learning": I am wondering how the test-error rate is calculated based on how the describe the simulation at the start? How do they ...
user394334's user avatar
1 vote
0 answers
26 views

Error propagation: How to sum errors over 2D grid?

I have a dataset with worldwide mass change data and their uncertainty from glaciers. Both have dimensions 720,360,45 with the first two dimensions 'i,j' (lat,lon) coordinates and the third dimension '...
yoniv1's user avatar
  • 11
2 votes
2 answers
59 views

Test and Train RSS in OLS model

I encountered the following true/false question: Given a train sample with $\ N $ observations and OLS model fitted on that sample, the RSS of the train sample will be less than or equal to the ...
bm1125's user avatar
  • 187
0 votes
1 answer
38 views

Interpolation of errors from model predictions over time-series

I have a regression model: ...
Squan Schmaan's user avatar
0 votes
0 answers
13 views

Difference in margin of error for boostrap and parametric approaches in survey

I'd like to know if using a bootstrap-derived margin-of-error for a simple survey is appropriate. I'm worried that the estimate is too small. I've pasted the code below for both approaches for context....
Statfan's user avatar
  • 165
1 vote
2 answers
67 views

In linear regression, does the formula for error contain the marginal expectation or conditional expectation?

In linear regression, let $\epsilon_i$ be the $i$th error term. Is the formula for $\epsilon_i$ $\epsilon_i = Y_i - E(Y_i)$ or $\epsilon_i = Y_i - E(Y_i | X_i = x_i)$? I have seen both definitions....
Iterator516's user avatar
1 vote
0 answers
42 views

How to compute relative error of multi-dimensional time-series?

I have written a python script that uses a variety of different integrators to simulate the gravitational N-body problem. I would like to compare the positions obtained from my simulation to the ...
user23358153's user avatar
0 votes
0 answers
12 views

How to handle seasonality when using relative errors

I am using a model that forecast predictions for DAUs (daily active users). The DAU dataset is seasonal, so I'm trying to figure out the right "error" function for my model. (The model I'm ...
nz_21's user avatar
  • 221
0 votes
0 answers
24 views

how to calculate overall RMSE accumulated during several processing steps

I have a digital terrain model (DTM) downloaded from NASA's SRTM dataset at a resolution of 1 arc second covering Spain and France. This has a stated RMSE of 9.73m [output 1] I projected this to ...
richjh's user avatar
  • 1
0 votes
0 answers
33 views

Statistical error analysis of vector data

Reaching equilibrium in a Monte Carlo simulation often refers to a state where the system has evolved sufficiently such that its statistical properties no longer change significantly with additional ...
user366312's user avatar
  • 2,195
4 votes
1 answer
188 views

how to compute clustering error properly

true = c(1,1,1,2) pred1 = c(1,1,2,2) pred2 = c(1,1,2,3) Suppose my dataset has two clusters, after using two clustering algorithms, one gives pred1 and the other ...
Simple's user avatar
  • 207
0 votes
1 answer
49 views

How do I combine the error when multiplying a number by a proportion [closed]

I wish to know how many seals are in an area. Seals have been counted in a portion of the area, once each month over multiple years. Separately, several seals in the area have been fitted with GPS ...
Roger James's user avatar
0 votes
0 answers
25 views

Error propagation. Simpler average errors

I am designing a lab practicum to study error propagation. Let's suppose I will measure $x \pm\varepsilon_x$ and $y \pm \varepsilon_y$, where $\varepsilon_x = \varepsilon_y = \varepsilon$ for ...
Daaviid's user avatar
  • 97
3 votes
1 answer
120 views

How does non-collapsibility and the lack of an error term affect coefficients in regression

I have read from here that in nonlinear models such as the logit and Cox, because of a lack of an error term, coefficients may be biased (typically towards zero) when covariates are omitted; I see how ...
Geoff's user avatar
  • 741
1 vote
0 answers
16 views

Drawing error bars on a position vs time chart where st. dev. is for vectors and not total position

I have the following data for a position versus time analysis. Note that the total distance is just the sum of the mean vector distances. I want to show the error bars on a chart of total distance vs ...
rdemyan's user avatar
  • 11
1 vote
1 answer
158 views

Symmetric AND Weight MAPE Calculation

I'm responsible to forecast a portfolio of consumer products on a monthly basis, and in calculating forecast accuracy, I'm lead to the MAPE (Mean Average Percent Error), which is useful, but has, ...
Mark J's user avatar
  • 11
1 vote
1 answer
83 views

Source of error in Linear Regression?

Suppose we are given n data points (observations) for random variable Y and variable X. We are to find regression equation of Y on X. As I’ve read these given values of Y (observations) are ...
Quorthon's user avatar
  • 107
0 votes
0 answers
61 views

Residuals and "error terms" in time series

I'm self-studying and I see “residuals” seems to be what is left, after we take away non-random components. So if we have additive decomposition : $$ Series = Constant + Trend\text{ }_t + Seasonality\...
Johannes Konrad's user avatar
6 votes
1 answer
297 views

OLS: do we test the residuals for normality *because* then the error terms can be assumed normal, too? Is there proof for this?

There are lots of resources out there that mix up residuals with errors, using the terms interchangeably, or saying "residual errors", or not acknowledging the existence of errors at all. (...
Reader 123's user avatar
1 vote
1 answer
56 views

Error of prediction from linear regression in R [closed]

I have an equation: $$ \large y = 0.243x + 0.145 $$ In the form: $$ \Large y = ax + b $$ I use it to predict $y$ when $x = 2$. To estimate the distribution around $\hat{y} = 0.631$ I need an estimate ...
Aaron Simmons's user avatar
0 votes
0 answers
20 views

Can you penalize based on difference between training and testing error?

Is it a valid or useful technique to penalize the model based on the objective function + abs(training error - testing error). The error would probably have to be scaled. This seems useful, and I am ...
BigMistake's user avatar
0 votes
1 answer
27 views

How report Standar error of measurement (SEM)? [closed]

How report standard error of measurement (SEM)? Always with "±" due is a range?
user avatar
0 votes
0 answers
21 views

in-sample error and the optimism

I'm currently reading p228 of The Element of Statistical Learning, which covers training error, in-sample error, and optimism. Let me quote some of the textbook contents as follows. The $Y^{0}$ ...
jason 1's user avatar
  • 311
0 votes
0 answers
27 views

Estimating number of occurrences of binomial tests

I have data representing a counting of the number of successes in a series of $n$-trial binomial experiments, however each experiment might have a different $n$ and is unknown. So, if I for example ...
scf's user avatar
  • 101
0 votes
0 answers
21 views

Peculiar silhouette score is reduced with more clusters

I am working with Python and a set of sensors. I want to use clustering on them. To do that I have used hierarchical cluster with average linkage and euclidean distance as: ...
slow_learner's user avatar
1 vote
0 answers
68 views

How to calculate the uncertainty of fitting parameter in a nonlinear model

I have a cost function which is: (F(X,B)-Y)*(F(X,B)-Y) F is my model, B are my fitting ...
MOON's user avatar
  • 173
5 votes
2 answers
449 views

Modelling the residuals of a model as a function of an external variable in order to assess its effect on the errors of the model?

I am working with field variables and telemetric variables. The dataset is composed of geographical locations for which i have both types of data. Amongst these field data, some are of interest, to be ...
Renaud Bied-charreton's user avatar
1 vote
0 answers
12 views

what does it mean to have slightly higher valid error rate, very high and similar valid error rate compared to train and test?

the dataset is split into 3 categories: train, test, valid. after training the error rates we see have these values. case 1 1% error on the training set. 5% error on training dev set. 5% error on the ...
ERJAN's user avatar
  • 111
1 vote
3 answers
207 views

Residuals in linear regression - variance and independence

Let's assume a simple linear regression $$ y = \beta_0 + \beta_1 x + \varepsilon $$ where $\varepsilon_i$ are independent and come from $\mathcal{N}(0, \sigma^2)$. We define residuals as $$ e = y - \...
thesecond's user avatar
  • 390
1 vote
0 answers
52 views

Compute Sample Size for Specific Margin of Error & Confidence Level

I have a population that is in size equals to 65,536. The population follows a uniform distribution and is composed of discrete, ordered integer numbers. From this population I wish to identify the ...
ex1led's user avatar
  • 111
0 votes
0 answers
102 views

Warning message: glm.fit: algorithm did not converge and diagnose perfect separation [duplicate]

model raises a warning: glm.fit: algorithm did not converge. One of the suspected causes of this error in the model could be due to perfect separation (source: https://www.bookdown.org/rwnahhas/RMPH/...
mutu's user avatar
  • 1
0 votes
0 answers
5 views

Should I sample or compute the innovation term in the recursive estimation OLS algorithm applied to time-series data?

I have time-series data $\{Z_{[t]}\} = Z_t, Z_{t-1}, ..., Z_1$ of length $L$, and I would like to estimate a model for the $\{Z_{[t]}\}$ so I can forecast $Z_{t+1}$ as a function of a number of its ...
Jxson99's user avatar
  • 664
0 votes
1 answer
49 views

Root Mean Square Error of the addition of two measurements whose RMS Error is known

I am working on a measurement system which tries to measure the distance between two values i.e $\Delta F=F_1-F_2$. Where $F_1, F_2$ are the values I actually measure. I have set up a Monte Carlo ...
bad_at_stats's user avatar

1
2 3 4 5
24