Hello,(adsbygoogle = window.adsbygoogle || []).push({});

Let [itex]X[/itex] be a set of [itex]N[/itex] lognormal prices (in dollars), meaning

[tex]\log(X) = Y \sim MN(\mu_Y , \Sigma_Y) ,[/tex]

i.e. the log of [itex]X[/itex] follows a multivariate normal distribution.

Imagine now that one wants to compute various quantiles for this set, e.g. 2.5%, 50% and 97.5\%, and does this by simulating 100k draws from the distribution above.

You then get say a [itex]100k \times N[/itex] matrix, and to get the total value you'd find these three quantiles for each of the [itex]N[/itex] prices, resulting in a [itex]N \times 3[/itex] matrix of quantiles, and then simply add up each of the three columns giving three numbers which represent the total quantiles for the whole set.

So, my question is:

How would one go about transforming the information for the quantiles for the whole set from log-dollars to dollars?

I am of course aware of the exponential function, so what I'm asking is how and where in this process do I use it?

------------------------------------------------------------------------------------------------------------------------------------------------------

My main idea is this:

The three final quantiles, 2.5%, 50% and 97.5\%, represent the log-quantiles of the set as a whole, say [itex] V_{2.5\%}, V_{50\%}[/itex] and [itex] V_{97.5\%}[/itex]. They also have their log/exp-counterparts, e.g. [itex] V_{50\%} = log(X_{50\%})[/itex].

Now, the difference between, for example, the 50% and 2.5%, [itex]V_{50\%} - V_{2.5\%}[/itex] could then be represented as

[tex] log(X_{50\%}) - log(X_{2.5\%}) = log\left(\frac{X_{50\%}}{X_{2.5\%}}\right),[/tex]

meaning that the log difference can be interpreted as log-proportional difference. Thus one could just exp and inverse this value and get

[tex] exp \left( log \left( \frac{ X_{50\%} }{ X_{2.5\%} } \right) \right)^{-1} = \frac{X_{2.5\%}}{X_{50\%}} [/tex]

which gives you the proportional difference of the quantiles in dollars, instead of log-dollars. With this information the 2.5% quantile can be obtained by multiplying [itex] X_{2.5\%}[/itex] with the mean of the [itex]N[/itex] prices.

Or am I way off here?

Any help is greatly appreaciated.

**Physics Forums | Science Articles, Homework Help, Discussion**

Join Physics Forums Today!

The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

# Quantiles of a log-multivariate-normal-distributed set.

Loading...

Similar Threads for Quantiles multivariate normal |
---|

A Defining a Symmetry Statistic to test for Normality? |

A Sum of independent random variables and Normalization |

I Sampling from a multivariate Gaussian distribution |

I Data from Normal D is Uncorrelated? |

**Physics Forums | Science Articles, Homework Help, Discussion**