Questions tagged [skewness]
Skewness measures (or refers to) a degree of asymmetry in the distribution of a variable.
730
questions
2
votes
1
answer
42
views
Should I normalise a sample drawn from a skewed population?
I am interested in the effect of the number of inhabitants of political entities on certain political behaviours within these entities. It happens that the number of inhabitants is not evenly ...
- 23
0
votes
0
answers
33
views
log transform left data in r
I am having trouble finding the transformation operation for left/negatively skewed data. The catch? All of my values are between 0 and 1. As such, trying the standard log10 transformation command ...
- 111
1
vote
1
answer
27
views
Why does a small shape parameter in the skew normal distribution invert the skewness?
I am using the rsnorm function from the fGarch R library. As one would expect, for shape parameter (xi in this implementation) that is sufficiently large (abs(xi) > 1), a histogram has a clearly ...
- 11
6
votes
1
answer
307
views
Vintage of this lower bound on skewness for positive data with given mean and sd?
It turns out there is a lower bound on the skewness $g_1$ of any strictly positive set of data having a given mean μ and standard deviation σ:
$$
g_1 > \sigma/\mu - \mu/\sigma.
$$
Although ...
- 2,117
0
votes
0
answers
20
views
zero-inflation analysis multilevel for continuos data (not count data)
I was trying to fit a multilevel model, but I discovered that my dependent variable is highly skewed and zero-inflated.
Individuals report 5 times a day for 7 days their level of paranoia and the ...
- 1
8
votes
4
answers
453
views
Estimate Box-Cox Transformation Lambda Using Skewness and Kurtosis
I would be interested in a method to find an appropriate Lambda parameter for the Box-Cox transformation based on only the skewness and the kurtosis of a given sample.
I.e, if the skewness and ...
- 425
1
vote
0
answers
41
views
How can a probability density distribution look negatively skewed on a graph but yield a positive skew result in Excel? [closed]
I am analyzing data from a young single mothers residential facility and I am seeing probability density plots that look clearly negatively skewed but Excel and calculates the dataset as being very ...
- 11
2
votes
1
answer
51
views
Why do the skewness and kurtosis formulae have powers of the variance in the denominator?
We calculate the variance as the centered 2nd moment $E[(X-\mu)^2]$.
So when it comes to the skewness and kurtosis, why are the 3rd and 4th moments divided by the 3rd and 4th powers of $\sigma$? Why ...
- 141
17
votes
5
answers
4k
views
Is a product that has 4.9 stars from ten customers better than one that has 4.5 stars from a hundred customers?
In many areas, we encounter a situation where we compare averages of highly skewed statistics using two unequally sized samples. Typically, this happens when comparing items in an online store. For ...
9
votes
3
answers
409
views
When does positive skew imply median<mean?
For a random variable $X$, the skew is defined as $S(X):={\frac {E\overline{X}^3}{(E\overline{X}^2)^{3/2}}}$, where $\overline{X}=X-EX$. It is often claimed that positive (resp. negative) skew implies ...
2
votes
2
answers
76
views
Logistic regression with highly left-skewed data for the independent variable
I am using the in-built GLM function in R to identify the model that best predict frogs' occupancy based on survey data. One of the independent variable (saturation) is highly skewed, as 36 of the 57 ...
0
votes
0
answers
37
views
Inequality regarding measure of skewness & kurtosis [duplicate]
The measures of skewness and kurtosis respectively are
$b_1=\frac{m_3^2}{m_2^3}$(skewness)
and
$b_2=\frac{m_4}{m_2^2}$(Kurtosis)
where $m_r$ is the central moment of $rth$ order. That is $m_r = \frac{\...
2
votes
0
answers
44
views
One-way repeated measures ANOVA with skewed response
We have an experiment with 102 individuals in total. We have an outcome $Y$ (which is a variable related to the structure of a given bone), and we want to know whether this variable $Y$ differs ...
- 359
2
votes
1
answer
108
views
How to normalise outputs of neural networks with different distribution?
I have a NN model that predicts 8 different variables. I use a multi-task learning approach, where I compute the loss between predictions and targets for each of ...
- 95
3
votes
1
answer
123
views
Zero variance but non-zero skewness
I was thinking of a hypothetical distribution where the mean(first cumulant) is non-zero, second cumulant(variance) is zero, and the third cumulant(skewness) is non-zero. The higher order cumulants ...
- 133