Questions tagged [density-estimation]

Estimation of probability density functions, whether by kernel density estimation, log-spline estimation or other methods.

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

state of the art in density estimation

I have seen density estimation methods which are pretty old. Specifically, I am referring to Parzen Window method. When I read the original Parzen's paper, I was amazed by it's beauty and I know that ...
2
votes
1answer
30 views

How to evaluate the accuracy of a probability distribution?

I've trained a Gaussian Bayesian Network. If I feed input values for the parent variables of my output variable, I get a normal distribution. How can I quantify the accuracy of this distribution when ...
0
votes
0answers
15 views

How can I estimate bivariate probability density for support restricted data?

I have a bivariate sample with the following kernel density estimation The issue is that there is actually a cutoff for log(Age) at about 2.5, so value greater than 2.5 has probability 0. The fitted ...
0
votes
0answers
12 views

Clustering timeseries subsequences (detection of modes)

I am working on a task that involves detecting different "clusters" of a timeseries signal. So basically I need to differentiate between "modes" (importantly, I do not know how many groups there will ...
0
votes
0answers
14 views

Have $(U,V)$ be a pair of Bivariate Gaussian variables with mean $0$, variance $1$ and $Cov(U,V) = p$ where 0 < ρ < 1 [duplicate]

Have $(U,V)$ be a pair of Bivariate Gaussian variables with mean $0$, variance $1$ and $Cov(U,V) = ρ$ where $0 < ρ < 1$ I'd like help finding the density of $U+V$ So far I have tried to use $...
0
votes
0answers
9 views

Is there an example reference for Density estimation using Triangular Kernel Function

Density estimation using Triangular Kernel Function
0
votes
0answers
12 views

Finding probability bucket which result in 90% correct classification

I have a dataframe of two columns, one of which contains probabilities of event X happening and the second column is whether or not X did occur as indicated by a 0,1. I would like to find the buckets ...
0
votes
0answers
42 views

Density Estimation Efficieny

My Question Let's say a set training samples like D from a discrete distribution like p(x) over a discrete variable vector like x is available. We don't have any prior knowledge about the form of p(x)...
1
vote
1answer
52 views

Estimate value with binomial distribution [closed]

We have some compound A diluted in a solution. In 200 trials, we find that when we mix $1$ $\mathrm{mm}^3$ our solution of A with some amount of some compound B, we get a reaction 185 times. How can I ...
1
vote
1answer
58 views

Density Estimation and Data Normalization

Is there any problem to first normalize data (for example, min-max one) then use kernel density estimation to get pdf of each sample? Thanks.
0
votes
0answers
25 views

Every point has the same probability?

I am reading "Pattern recognition and machine learning" by Cristopher Bishop. In Chapter 2.5.1 "Kernel density estimator", there is written that: Let us suppose that observations are being drawn ...
0
votes
1answer
32 views

Bayesian approach: ignoring the denominator leads to the conditional density equaling the joint density? [duplicate]

I know there are a lot of questions here about ignoring the denominator in a Bayesian approach, but I don't think mine is a duplicate of any of them. I am reading the book "Pattern recognition and ...
2
votes
1answer
42 views

What is the difference between probabilistic forecasting and quantile forecasting?

A probabilistic time series forecast outputs the entire distribution of the forecasted values for a given time point, instead of just a mean or a point forecast. A quantile forecast is a forecast ...
0
votes
0answers
11 views

Piecewise Pareto Distribution

What are the best practices for Piece-wise Pareto Distribution or maybe Pareto Mixture Model(?). Example: $x\in [0, 1) \Rightarrow \alpha=0.1$ $x\in [1, 10) \Rightarrow \alpha=0.5$ $x\in [10, 100) ...
0
votes
0answers
33 views

To what distribution it's similar? Looks like an exponential but it's not

To what distribution it's similar? Looks like an exponential but it's not. It's seems to have a property, that if I zoom it (xlim) then each time it has the same ...
0
votes
2answers
70 views

Can Kernel Density Estimation estimate an Exponential Distribution?

Can Kernel Density Estimation estimate an Exponential Distribution? I tried to performed to make experiments with various kernels like: "gaussian" and "exponential", but performance seems to be very ...
0
votes
0answers
12 views

Comparing Density Plot height (MachineLearning Classification)

I am working on a binary Machine Learning classification problem. My classifiers are really performing poorly because distribution of the 1 class is very similar to 0 class (dataset is imbalanced, 1 ...
1
vote
0answers
33 views

Substitution for unknown true density in 'Density Estimation Trees'

I'm having a hard time understanding parts of the derivation of the objective function for Density Estimation Trees (reference below) regarding the loss function. Taken from the article (Sec. 3.1): ...
0
votes
0answers
32 views

Non-parametric estimation

This is some sort of an initial question may be I'm asking which may not have a fixed, straightforward answer. But this is very unclear to me. I have come up with some parameter estimation methods for ...
0
votes
0answers
28 views

Conditional Density Estimate Loss. Why the double integral?

I read RFCDE: Random Forests for Conditional Density Estimation. Just like it sounds, these folks trained random forests for making conditional density estimates. At inference time, density estimates ...
0
votes
0answers
23 views

bandwidth setting for density comparison

I intend to compare a list of density distributions. Following one published paper (with similar type of data and same objective), I learned that I have to use Gaussian kernel density estimator, and a ...
3
votes
1answer
57 views

Properties of Kernel Density Estimators

Given Let $X \in \mathbb{R}$ be a real-valued random variable with theoretical probability density function (pdf) $f(x)$ and corresponding cumulative distribution function (cdf) $F(x)$. Let $X_1, X_2,...
0
votes
0answers
22 views

Integration of Kernel and density product

Im considering kernels of the form $$K_s(u) = A(s)k_s(u)I[\lvert u \lvert \leq 1]$$ and $$k_s(u) = (1-u^2)^s$$ with $r$'th derivative $$K_s^r(u) = A(s)\frac{d^r k_s(u)}{du^r}I[\lvert u \lvert \leq ...
0
votes
0answers
19 views

Expectation of derivative of kernel density estimator

I am trying to calculate the expectation of the $s$'th derivative of a kernel density estimator. This problem arises naturally when trying to estimate the derivative of a density, because one approach ...
1
vote
0answers
47 views

Is There an Optimal Bandwidth for Kernel Density Estimation at One Point

I am trying to estimate the density of a 2D distribution using KDE, but I only care about the accuracy right at the origin. I have read about methods to estimate the optimal bandwidth matrix when you ...
0
votes
1answer
19 views

Function dependent to nth order on a sample

From this paper: ... every multivariate density estimator that is in any reasonable sense nonparametric may be written in the form: $$ \hat{f}(y) = \frac{1}{n}\sum_{i=1}^n K_n(x_i, y) $$ ...
2
votes
1answer
309 views

Difference between function and distribution?

I know this is dumb question, but i am confused to understand it. I have constant function as $$y = exp(-x^2)$$ This also represent gaussian distribution with mean zero and variance of 0.5 Now if ...
0
votes
0answers
55 views

Are vanishing bias and variance enough for pointwise consistency for KDE-based estimation?

Question: Is the condition that asymptotic bias and asymptotic variance goes to zero for infinite samples sufficient to guarantee the pointwise consistency of an estimator based on plug-in kernel ...
1
vote
0answers
70 views

Can kernel density estimation be used to estimate a radial probability density?

I am trying to estimate the PDF of the radius of points distributed in a 2D plane. The points are distributed like this: I can produce a histogram of the radius data which looks like this: I want ...
1
vote
1answer
120 views

Credibility evaluation - how to model conditional continuous density from multiple variables of various types?

I recently got dataset for 37000 households with declared income and a few dozens of other variables of various types: continuous, discrete, binary. The task is to automatically (unsupervised) ...
0
votes
0answers
66 views

Weighted kernel density

I would like to produce a 3-d plot based on density of 2-d data. This can be achieved for example in R using the kde2d function from the ...
0
votes
2answers
133 views

GLMM species count data with transects

I am trying to create a GLMM model which explains differences in abundance/count of three species of scorpion around a field reserve in different forest types. -I have 7 trails in different forest ...
0
votes
1answer
155 views

How to get a density from a forecast with prediction interval

Some reproducible code to have in your environment a time series and a possible forecast: ...
2
votes
0answers
31 views

Choosing Gaussian PDF basis bandwidth depending on number of bases and range of data

Summary (details below!) I have a basis expansion of $m$ (univariate) Gaussian PDFs to model the density of a sample $X$. The means of these PDFs are spaced equidistantly through the domain of $X$ ...
0
votes
0answers
10 views

Principled estimate of joint PDF given marginals and first and second-order statistics

(Trying to solve the same underlying problem as this.) Let's say we know the marginal probability density functions $p_i(x)$ of a set of zero-mean random variables $\{X_i\}_{i=1}^N$ as well as the ...
1
vote
3answers
92 views

How long is a distribution considered normal?

I have a dataset of metric distances ($n=5800$) and plotted those as a histogram. My initial thought was that this distribution looks normal. But after performing a Shapiro Wilk test and plotting the ...
0
votes
0answers
13 views

What are the pitfals of using kernel density estimates to infer about the shape of an underlying distribution

I know that we cannot simply infer from the shape of a histogram to the shape of the underlying distribution, as the shape of the histogram is influenced by the choice of the intervals (Assessing ...
0
votes
1answer
23 views

Is the inverted pi a standard symbol for d-separation?

I've seen the inverted pi symbol earlier but found it hard to find an explanation for it. Found it again in the context of d-separation between Xu and Xv here: https://en.m.wikipedia.org/wiki/...
0
votes
1answer
35 views

Multiplying two event with probability density function, is it possible?

in my exercise, $X$ is the size of a tree trunk, and $X$ follows a normal distribution $\mathcal{N}(9,0.4)$, we want to know $P(8.8\le X\le11.2)$ So I though that I could do this: $P(8.8\le X \le11.2)...
1
vote
1answer
49 views

How to determine Density Value of a Z-Score?

In this paper the authors are demonstrating the conversion of Ordinal data to Interval data. In the Step No. 7 on Page 360, they are talking about Density values of corresponding Z-values. They are ...
2
votes
0answers
98 views

Finding probability of a point using bivariate copula density

I have a data in the form $\textbf{N} \times 2$. I am using bivariate copula to model the joint density of this distribution. Firstly, I fit 2 marginal distributions independently on each column of ...
0
votes
1answer
189 views

MSE of Kernel Density Estimator

Erlang Kernel is used for density estimation. By using this estimates are pretty close to the real density on graph on the other side MSE is very large. But Author of Erlang Kernel stated that it will ...
2
votes
0answers
71 views

How to aggregate histograms for density estimation

Within a very large sensor network, each node does take measurements derived from a fixed number of samples taken at a high frequency from an instrument. The number of measurements send to an ...
2
votes
1answer
791 views

A Gaussian Mixture Model Is a Universal Approximator of Densities

When discussing the concept of mixtures of distributions in my machine learning textbook, the authors state the following: A Gaussian mixture model is a universal approximator of densities, in the ...
1
vote
0answers
42 views

Truncated Non-Parametric Density Estimation

I have cross-sectional data for individuals, $i$, and years, $t$. I need to estimate the density of a random variable, $X_{it}$, which is equal to $X_{it} = A_{it}/B_t$. My data is truncated in the ...
0
votes
0answers
63 views

Density curve in R - AUC bigger than 1

correct me if I'm wrong but I was expecting the area under the curve should be 1 for a probability density function. Can anybody explain why it's not always the case when using the ...
0
votes
1answer
628 views

Bandwidth parameters in multivariate KDE using scipy.stats.gaussian_kde

I am working on a project which involves implementing in Python two different density estimation functions over multivariate data; one using N-d histograms and the other using kernel density ...
2
votes
0answers
10 views

How to estimate density given similar instances?

I'm trying to estimate the probability distribution of a random variable $X$ given values of $N$ parameters/qualities $Q_1, Q_2, ..., Q_N$, and historical data. The two extremes in estimation would ...
4
votes
1answer
107 views

Why isn't mean integrated square error (MISE) probability-weighted?

We often try to minimize the MISE of a KDE: $\text{E}_{\mathbb{P}^n}[\int (\hat{p}(x) - p(x))^2 dx]$. Why don't we instead try to minimize $\text{E}_{\mathbb{P}^n}[\int (\hat{p}(x) - p(x))^2 p(x) dx]$,...
4
votes
1answer
86 views

Assessing the quality of a density estimate without knowing the true density

Assume that for a univariate sample $X$ you have a density estimate by a model you developed (in my case it's a density forecasting model) and you want to assess the quality of your models estimate. ...