## Abstract

A classical result states that the sample variance of a standard Gaussian sample has the chi-square distribution. In this note, a partial reverse of this result is proved for independent infinitely divisible random variables *X*
_{1},…,*X*
_{
n
},*n*≥2. If *n*≥3,\(\mathbb {E}X_{1}=\cdots =\mathbb {E}X_{n}\) and the random variable \(n{S^{2}_{n}}=(X_{1}-\overline {X})^{2}+\cdots +(X_{n}-\overline {X})^{2}\) where \(\overline {X}=(X_{1}+\cdots +X_{n})/n,\) has the chi-square distribution with *n*−1 degrees of freedom then *X*
_{1},…,*X*
_{
n
} are Gaussian random variables with \(\mathbb {E}(X_{1}-\mathbb {E}X_{1})^{2}=\cdots =\mathbb {E}(X_{n}-\mathbb {E}X_{n})^{2}=1\). In the case *n*=2, the random variable \(2{S^{2}_{2}}\) has the chi-square distribution with 1 degree of freedom if and only if *X*
_{1} and *X*
_{2} are Gaussian random variables with \(\mathbb {E}X_{1}=\mathbb {E}X_{2}\) and \(\mathbb {E}(X_{1}-\mathbb {E}X_{1})^{2}+\mathbb {E}(X_{2}-\mathbb {E}X_{2})^{2}=2\).

This is a preview of subscription content, access via your institution.

## References

Hogg, R.V. and Craig, A.T. (1970).

*Introduction to mathematical statistics, 4th edn.*Macmillan Publ. Co. Inc., New York.Kruglov, V.M. (1970). A note on the theory of infinitely divisible laws.

*Theory Probab. Appl*.**15**, 319–324.Kruglov, V.M. (1972). Integrals with respect to infinitely divisible distributions in a Hilbert space.

*Math. Notes***11**, 407–411.Kruglov, V.M. and Antonov, C.N. (1982). On asymptotic behaviour of infinitely divisible distributions in Banach space.

*Theory Probab. Appl*.**27**, 626–642.Kruglov, V.M. (2003). Normal and Poisson convergence.

*Theory Probab. Appl*.**48**, 392–398.Kruglov, V.M. (2010). A characterization of the Poisson distribution.

*Stat. Probab. Lett*.**80**, 2032–2034.Kruglov, V.M. (2012). A characterization of the convolution of Gaussian and Poisson distributions.

*Sankhyā Series A***74**, 1–9.Kruglov, V.M. (2013). A characterization of the Gaussian distribution.

*Stoch. Anal. Appl*.**31**, 872–875.Loève, M. (1977). Probability theory, I, 4th edn. Springer-Verlag.

Rossberg, H.-J., Jesiak, B. and Siegel, G. (1985). Analytical methods of probability theory. Berlin, Akademie-Verlag.

Ruben, H. (1974). A new characterisation of the normal distribution through the sample variance.

*Sankhyā Series A***36**, 379–388.Ruben, H. (1975). A further characterisation of normality through the sample variance.

*Sankhyā Series A***37**, 72–81.Ken-Iti, S. (1999).

*Levy processes and infinitely divisible distributions*. University Press, Cambridge.Ssörgő, A. (2005). A probabilistic proof of Kruglov’s theorem on the tails of infinitely divisible distributions.

*Acta Sci. Math*.**71**, 405–415.Yakymiv, A.L. (2005). Probabilistic applications of Tauberian theorems. Leiden-Boston, VSP.

## Author information

### Authors and Affiliations

### Corresponding author

## Additional information

Research supported by the Russian Scientific Foundation, project 14-11-00362.

## Rights and permissions

## About this article

### Cite this article

Golikova, N.N., Kruglov, V.M. A Characterisation of the Gaussian Distribution through the Sample Variance.
*Sankhya A* **77**, 330–336 (2015). https://doi.org/10.1007/s13171-014-0060-5

Received:

Revised:

Published:

Issue Date:

DOI: https://doi.org/10.1007/s13171-014-0060-5

### Keywords and phrases

- Gaussian distribution
- chi-squared distribution
- distribution theory

### AMS 2000 subject classifications

- Primary 60E07
- Secondary 62E10