Joint probability with housing stock data

38 Views Asked by At

I have a question about probability. Let’s say we have 100 homes of different ages, types and insulation levels, distributed as per the table below.

Housing stock data

How do I determine how many homes would fall into this category: “Pre 1900-1929, 3 bed house, Solid walls, No insulation, Double glazed, Gas”

A simple probability formula gives this answer: =(70/100) * (60/100) * (50/100) * (60/100) * (70/100) * (90/100) = 7.9% = 8 homes

But this seems too low, given I know there will be several cross-overs in house types and a lot of homes will fall into the category I selected. Is there a more appropriate way of working it out?

Thanks

1

There are 1 best solutions below

0
On

You need more detailed data if you want to evaluate the proportion of houses with specific characteristics.

Lets say you have information about the characteristics windows (single glazed [$sg$], double glazed [$dg$]) and heating (Gas [$G$], No Gas [$nG$]).

$\begin{array}{|c|c|c|c|} \hline & sg&dg \\ \hline G &20 &70 &90 \\ \hline nG&10 &0&10 \\ \hline &30 &70&100 \\ \hline \end{array}$

Using your method the proportion of houses with single glass [$sg$] and gas heating [$G$] is $\frac{90}{100}\cdot \frac{30}{100}=27\%$. But here the real value is $20\%$.