Questions tagged [power-law]

A power-law is a function that increases proportionally to a power of its argument (ax^b). Often seen in fitted relationships or in densities (power-law distributions).

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

Difference between Power law distribution and Exponential distribution?

What is the difference between Power law distribution and Exponential distribution? They both look similar!
user avatar
0 votes
0 answers
90 views

Cumulative distribution function in powerlaw distribution

I am trying to fit the power law distribution to continuous sample (x) which has a sample size n = 360 I'm using the methodology proposed by Aaron Clauset in the python package power-law: A Python ...
  • 1
1 vote
0 answers
9 views

Ranking log-linear distributions and the Lucas numbers

The wikipedia page on rank-size distributions claims: "When any log-linear factor is ranked, the ranks follow the Lucas numbers, which consist of the sequentially additive numbers 1, 3, 4, 7, 11,...
1 vote
1 answer
75 views

Convergence of Percentile in Power Law

I have a probability distribution, that in its tail follows a power law. I've noticed, while I was simulating samples, and determining parameters experimentally, that as I increase the value of a ...
  • 243
1 vote
1 answer
332 views

Maximum likelihood estimator for power law with negative exponent

Background I have data that roughly follows a power law with a negative exponent (up to a point; also, the parameters of the "fit" were just guesstimated by eye as a demonstration): Now I ...
  • 105
0 votes
0 answers
19 views

Exponentially distributed bins for viewing and analysing the tails of a histogram

Is there a rule for visualising and analysing density estimates of heavy tailed observations on a log-log scale? For example, equispaced bins and exponentially distributed bins: ...
  • 43
0 votes
0 answers
17 views

Optimal new feature selection that reduces uncertainty on a fit

I have a dataset that contains a couple of hundred results from a computational task related to plasma physics. The computation time of the task is, due to the high complexity of the underlying ...
  • 1
2 votes
1 answer
60 views

Correct GLM or NLS to model exponential model with response variable with positive and negative values

I have been struggling to find the right way to model this dataset, this is a Data Frame with the dataset: ...
0 votes
1 answer
70 views

What are the theoretical assumptions of a power curve versus an exponential or quadratic one?

I have a set of bivariate biological data that has a clearly non-linear distribution. I have found that a power curve best fits the data under several metrics (e.g., AIC, residuals versus fits plot, ...
0 votes
1 answer
171 views

Do stock price changes follow Pareto Distribution? [closed]

I calculated the distribution of the stock price daily ratios. The ratios are multiplicative, $d_t=p_{t} / p_{t-1}$. As far as I know the distribution should look like Power law distribution (Pareto ...
0 votes
1 answer
35 views

How do I calculate a proper sample size to get a suitable power of Kolmogorov-Smirnov test with an underlying Zipf distribution?

I have been trying to develop to calculate the sample size to maximize the power of KS test (±0.8) on an underlying Zipf distribution. I have tried estimating the power by performing simulations: <...
1 vote
1 answer
198 views

Using AIC vs Likelihood Ratio test for comparing Lognormal and Powerlaw distributions

I am interested in comparing whether a lognormal or a power law are a better fit for a given set of data. Both distributions have been fit using MLE, with $x_{min}$ determined using KS-minimization a ...
  • 31
2 votes
1 answer
39 views

Reading on "Economic nowcasting" for a probabilist

I am about to attend an interview for a postdoc in "economic nowcasting". I am work mostly in probability, particularly complex networks and random graphs. I am looking for some good reading,...
  • 121
1 vote
0 answers
24 views

Why if a random variable is power law distributed I should consider ln N (with ζ=1) in place of sqrt(N)?

In this paper I don't get Preposition 2. In particular firm size can be described as: $$ \DeclareMathOperator{\E}{\mathbb{E}} dS_{it}=0S_{it}+\sigma S_{it}\; dW_{it} $$ or according to the paper $...
  • 69
1 vote
0 answers
14 views

Symbol interpretation in non-constant power law density

I am reading Barabasi book on network science. I am struggling to interpret the density formula for a power-law degree distribution with low-degree saturation and high-degree cut-off (http://...
  • 2,343
8 votes
1 answer
337 views

Definition of heavy-tailed distribution

I'm reading about heavy-tailed distributions, the definition states that: The distribution of a real-valued random variable $X$ is said to have a heavy right tail if the probabilities $\mathbb{P}(X &...
4 votes
1 answer
130 views

Why is a Fréchet distribution slowly varying, and what is the intuition behind it?

The Fréchet distribution: $$\Phi_\alpha(x)=\begin{cases}0 & & x\leq 0,\\[6pt]e^{-x^{-\alpha}} & & x>0,\end{cases}$$ is regularly varying as stated here (page 19): It is not ...
2 votes
1 answer
88 views

Why does the tail of a Fréchet distribution decay as a power law?

The Fréchet distribution: $$\Phi_\alpha(x)=\begin{cases}0,\; x\leq 0\\e^{-x^{-\alpha}},\; x>0\end{cases}$$ shows a power law decay at the tail (survival): $$1- \Phi_\alpha(x) = 1 -e^{-x^{-\alpha}}\...
4 votes
1 answer
219 views

Student's t as a power law distribution

I'm currently reading about power laws and I have came across an answer stating: The density function of a Student's t-distribution with $n$ degrees of freedom is: $$f(x) \sim (1 + x^2 / n)^{-(n+1)/2}...
3 votes
1 answer
538 views

How can this CDF be decreasing? (power-law)

I am reading the first vignette of the powerRlaw R package. It uses the moby dataset, which is an array of word frequency. Each element is a word, and the value if ...
3 votes
1 answer
2k views

Difference Between SciPy.Stats.powerlaw and powerlaw package

I am trying to simulate random variables that are power law distributed based on my understanding of the definition in this Wikipedia article and several other resources where the consensus is that a &...
  • 39
1 vote
0 answers
53 views

Sample size determination for power law

I am looking for a method to estimate a mean when the underlying samples are drawn from a skewed distribution. Concretely, I have a population $P$ whose pdf largely follows a power law, where only a ...
  • 41
1 vote
2 answers
148 views

What is the mathematical meaning of when two variables retain a non-linear relationship even after log transformation?

I am working with a set of biological data drawn from a sample of species, specifically a skeletal measurement and body size, with the intent of calculating a regression line to try to predict the ...
1 vote
0 answers
78 views

Is Prices law a consequence of Benford's law?

Background Price's law says this: In an company, half of all value is created by the square root of the people. I'm sure Dr. Price looked at more than the median, but I suspect this was how the ...
  • 8,428
1 vote
1 answer
75 views

Fit a function whose asymptotics is known

Consider the experimental blue curve in the figure that follows It turns out that I know the expression for the tails (-they are power laws- red and green curves) and I want to fit the curve using ...
  • 111
2 votes
1 answer
59 views

Help with Power Curve MCMC

I'm trying to analyze some $D$ non-logistic cumulative data in a time series, bounded below by 0 and unbounded above. Splitting data into $W$ time windows of $d$ days, I know each window can be ...
3 votes
2 answers
380 views

Sampling distribution of the mean of the discrete-power law distribution

For a certain problem I wish to generate random integers $k$ so that their distribution follows $p_k \sim k^{-\alpha}$ for $k \geq k_{\text{min}}$, $k_{\text{min}} > 0$. I am following the ...
  • 603
0 votes
1 answer
47 views

In the poweRlaw package, is the location parameter estimated for a lognormal distribution the $median=\exp(\mu)=\theta$ or $\mu$?

I have a set of graphs (each with the same nodes but with edges' weights defined by different research subjects) for which I would like to report statistics. One of these statistics is the betweenness ...
1 vote
0 answers
182 views

Why do poweRlaw and fitdistrplus differ in fitted lognorm parameters

I am trying to evaluate whether a power-law fit is appropriate for some data of lake areas that we have, and whether the theoretically supported alternative of the log-normal, at least at the tails, ...
  • 31
1 vote
0 answers
37 views

Appropriate Data Analysis when Criterion has Heavy-Tailed Distribution

I have a data set where my independent variables (i.e., personality assessment scores) are continuous and follow a normal distribution. The criterion, sales performance, is heavy-tailed and follows a ...
  • 71
0 votes
0 answers
41 views

Is there a named distribution with the property $P(X>10^k) = p^k$?

If I'm doing my math correctly, the exponential distribution has the property $P(X >k) = p ^ k$ (with $p$ conventionally written as $e ^ {-\lambda}$). I'm wondering if there is a different ...
  • 2,498
0 votes
0 answers
32 views

Why Are Bell Curves Used for Performance Appraisals?

Why do organizations that stack-rank employees force-fit employee rankings to a bell curve? Is there any theoretical or emperical evidence to support a normal or near-normal distribution of ...
0 votes
1 answer
2k views

Cut off long tail of distribution (no matter how long is the tail) [closed]

One of the algorithms that we are working on is designed to report the time it takes for people to finish the onboarding process in an app. We need an algorithm to eliminate the long tail of a ...
2 votes
0 answers
632 views

How to perform the Kolmogorov-Smirnov test on the truncated power-law distributed data?

Recently, I read papers that perform power-law fitting on their empirical data (estimate the alpha), some of them report corresponding p-value for the Kolmogorov-Smirnov test, but many of them do not. ...
  • 185
1 vote
0 answers
23 views

Powerlaw Distribution and old nodes [closed]

I am working on the size of discussions (number of replies, replies to the replies) on Twitter data, and I am observing Power-law distribution for small and medium-size discussion but it is not valid ...
0 votes
0 answers
74 views

Rank-size log-log plot inflection: drooping tail of power law

Most of my data seems nicely to fit a power law but with a "drooping tail", which I believe is quite common, although in this case the drop-off is quite steep. I have two related questions if I may ...
  • 235
2 votes
0 answers
23 views

How to Select the Largest Y Values for {X,Y} Pairs, for a Pareto-Distributed Dataset, for a Meaningful Fit?

First, apologies for the inelegant question. Second, on to the question: Background Information: I study impact craters, and the size-frequency distribution (number vs diameter) of impact craters ...
0 votes
0 answers
109 views

Random Samples from Spliced Distribution

I am studying Clauset, Shalizi, and Newman, Power Law Distributions in Empirical Data (preprint available here) in R. Packages used: ...
1 vote
0 answers
73 views

How can I calculate the exponent (parameter or tail index) of the Pareto distribution associated to the following paperclips experiment?

The experiment is the following: take a pool of paperclips for which in each round we follow these steps: 1) choose two paperclips randomly 2) if they are not linked with each other then link them ...
  • 111
1 vote
0 answers
129 views

Finding the exponents of a multiple power law: is linear regression valid?

I want to fit a multiple power law equation of the form $y = {x_1}^{\alpha_1} {x_2}^{\alpha_2}$ where I have many examples of $y, x_1, x_2$. (Note there is no intercept.) Is it possible for me to ...
3 votes
1 answer
898 views

Discrete Pareto Distribution vs Zipf Distribution and Power Law vs Zipf Law

I need to get a simple, but clear idea of Discrete Pareto Distribution vs Zipf Distribution and Power Law vs Zipf Law. (Are they similar/ how they relate to each other.) Wikipedia definitions do not ...
1 vote
0 answers
315 views

How are these two power law fitting glm models different?

I have some data that I thought I'd try fitting with a power law (in R). ...
  • 245
1 vote
1 answer
1k views

Using log-log graph to find equation of power law relationship?

I have a set of data that I think forms a power law relationship, however I am struggling to work out the equation of the relationship. Here is a subset of the data I am working with: ...
  • 225
1 vote
0 answers
197 views

Sum of powers of standard normal random variables

Context: While trying to teach the Central Limit Theorem I thought it would be a good idea to show a case where it breaks down. Question: Consider the sum of increasing powers of standard normal ...
  • 753
1 vote
1 answer
223 views

How well is a power law distribution described by the first four moments?

For a normal distribution, the first two moments (mean and variance) are sufficient statistics for the entire distribution. Suppose I have a power law distribution, and I have data on the first, ...
  • 678
3 votes
1 answer
566 views

Is the truncated power law a heavy-tailed distribution?

A heavy-tailed distribution is often defined as a distribution with a tail that is not exponentially bounded. A truncated power law (or power law with exponential cut-off) is a distribution that ...
  • 133
3 votes
3 answers
832 views

Most accurate way to express the income of the top 1% in a power distribution?

I have an array of 50,972 household incomes for a small metro area, and I want to measure how much the so-called top "one-percent" make (e.g., the 99th percentile). As you'd expect, it's a power curve:...
10 votes
2 answers
755 views

Is KS test really appropriate when validating a power law/estimating power law parameters?

I'm attempting to find out whether some highly skewed data are drawn from a power law distribution, following the popular paper by Clauset, Shalizi and Newman, 2009. Clauset et al. use the Kolmogorov-...
  • 203
1 vote
0 answers
53 views

Connection between logit model and power law distribution

I am wondering whether it is appropriate to make a connection between the use of a logit model and a power law distribution. My dependent variable is categorical and ordinal, therefore I am fitting a ...
2 votes
1 answer
930 views

Calculate Zipf-Mandelbrot parameters from distribution

I am fetching trending topics from social media where the frequency of likes is said to follow a Zipf-Mandelbrot distribution; i.e., some of the posts will have a high number of likes and some other ...