A hedonic model of prices is one that decomposes the price of an item into separate components that determine the price.
A simple, and common, example is that the price of a house may depend on its size, its location and other factors. It possible to construct a better model (especially when trying to construct a predictive model) by separating these factors.
A hedonic model does not necessarily separate all the factors that could be separated, only those that affect the usefulness to a buyer of what is being sold. A good model should also separate out other factors: for example, a model of house prices should also separate out interest rates.
One use of hedonic models is to adjust measures of inflation. The difference this makes is most significant for products that are not directly comparable with those that were sold in the past because technology has improved. Consider an entry level personal computer bought today compared with one bought ten years ago: the price is not greatly different, but it is much cheaper than the price of a similar computer bought ten years ago.
The adjustment that is needed is more complex than it may appear, as it should reflect the increase in usefulness (strictly speaking utility) of a product to customers. A computer that is ten times more faster with ten times as much memory etc. is not necessarily ten times more useful to a customer, and it is the increase in usefulness that is being measured.
The actual modelling can be quite complex, as shown by this description of the calculation of a hedonic price measurements for clothes driers.
Hedonic price factors are not usually of much importance to investors except when doing detailed modelling of product prices. This will only occur when doing very detailed models or when looking at prices that can naturally be decomposed into hedonic elements (such as house prices).