To reduce user-perceived latency in retrieving documents on the world wide web, a commonly used technique is caching both at the client's browser and more gainfully (due to sharing) at a proxy. The effectiveness of Web caching hinges on the replacement policy that determines the relative value of caching different objects. An important component of such policy is to predict next-request times. We propose a caching policy utilizing statistics on resource inter-request times. Such statistics can be collected either locally or at the server, and piggybacked to the proxy. Using various Web server logs, we compared existing cache replacement policies with our server-assisted schemes. The experiments show that utilizing the server knowledge of access patterns can greatly improve the effectiveness of proxy caches. Our experimental evaluation and proposed policies use a price function framework. The price function values the utility of a unit of cache storage as a function of time. Instead of the usual tradeoffs of profit (combined value of cache hits) and cache size we measure tradeoffs of profit and caching cost (average allocated cache portion). The price-function framework allows us to evaluate and compare different replacement policies by using server logs, without having to construct a full workload model for each client's cache.