System security and power quality is important in today's society and the ability to regulate and balance production and consumption is crucial for any power system. More and more penetration of intermittent production in power systems increases the need for regulation capability and the importance of capacity reserve markets where capacity used for regulation is procured and secured increases too. Several types of regulation mechanisms are used in a power system, which creates the possibility of several different capacity reserve markets with significant prices. A producer participating in these markets must decide how his limited production capacity should be used taking these markets and other physical power markets into account. A method for finding true costs for capacity reserve supply and for bidding in sequential capacity reserve markets is presented in this report. The method is based on a mixed-integer programming model and work has been done to create and formulate a suitable model. The modeling is implemented with the programming language AMPL and is an optimization model that maximizes total profit on several markets subject to market prices and market obligations for a set of production units. The model is then used to highlight some of the fundamental mechanisms and charactheristics in the markets and to illustrate the bidding method for a price-taking producer in perfect markets.Price uncertainty in future markets has a large impact on the results from the method and a model version where price uncertainty is included for the spot market is compared to a version where price uncertainty is not included. The reason for this comparison is that hourly spot price forecasts used for short-term production planning in Norway today doesn't consider price uncertainty. The versions are compared for bidding in one capacity reserve market for a number of market clearings where prices for the spot market in the model are taken from real spot price forecasts and real spot price outcomes. It shows that inclusion of price uncertainty gives better bids, but also that adjusting bids to account for price uncertainty can give good results from a model that doesn't explicity include this uncertainty. The method can in any case calculate valid bids for capacity reserve market solutions that exist today where costs and opportunity costs from all relevant markets can be accounted for. The limitations of the method is mostly connected to what it is possible to describe with mixed-integer programming and the computational efforts and calculation times mixed-integer programming models require.