Hi, hyang,
I think you might want to think about a metric you want to optimize, such as the monthly/weekly revenue of this new product, or number of items sold per month/week, while taking the price as an important factor. Then, from the historical sale history, you
can train a model and see how price is impacting the metric. The model might give you some insight of the relationship between the price and the metric.
Alternatively, if you do want to directly determine the price, you can build a regression model, where the target is the price, the variables might include features of products, how long this product has been on market, the demand of this product, whether
it is seasonal, holiday related product, etc.
Hope at least it can give you some hint at least.
Thanks.
Hang