**(This is a draft and truncated version - for final and full version, see**

Concise Encyclopedia of Biostatistics for Medical Professionals)

Concise Encyclopedia of Biostatistics for Medical Professionals)

**bivariate Gaussian distribution**

A bivariate Gaussian distribution, better known as bivariate normal distribution, is a special type of

**bivariate distribution**which follows a bell type shape (Figure B.7a). This figure has 3 axes. In the context of systolic and diastolic BP example in Table B.4 (but with ungrouped data), the bottom two axes depict systolic (

*x*) and diastolic levels (

*y*) and the vertical axis depicts number of subjects. Note the symmetry of this shape and equal standard deviations of

*x*and

*y*in Figure B.7a. In this case, you will see same shape no matter which side the plot is viewed from. The frequency is highest (peak of the distribution) when

*x*and

*y*values are at the middle of their ranges and the decline of frequencies on either side of these values follow the same pattern. The peak is such that

**kurtosis**is zero. ... ...

**For final and full version, see**

Concise Encyclopedia of Biostatistics for Medical Professionals

Concise Encyclopedia of Biostatistics for Medical Professionals

**Figure B.7:**(a) Bivariate Gaussian distribution with equal standard deviation of

*x*and

*y*; (a)Bivariate distribution of data in Table B.4; (c) Bivariate Gaussian distribution with unequal standard deviations