Agents operating in the highly competitive and volatile financial markets are forced to opt for top-quality risk management procedures.

The history of those markets abounds in examples of strong turbulences, most of which have been unexpected. Such unexpected if very real shocks have taught many a financial institution a painful lesson. Having learned from the past, we must continue research with a view to the future. Stress testing is a helpful tool in this research. It aims to analyse the impact, upon a company´s financial position, of a potential change in selected risk factors relevant to its activity, which may - although again may not - occur. Stress testing is one of the most important techniques in risk management. It can be defined as a set of methods utilised chiefly by financial institutions in order to determine their own immunity to extraordinary, if possible unfavourable developments in the market (market shocks). Shock resistance tests, or stress tests as the author calls them, are primarily used as a complementary tool in determining the value at risk. The paper looks further into the application of the tests; the suggestion is that they should be applied to the areas where other risk management models may fail to identify all the threats. Thus, in a sense, stress testing verifies and evaluates the assumptions of other models such as VAR or EAR, while on the other hand exploring the areas not covered by those models. Stress testing is also used to indicate the role of specific assumptions as to correlation and price variability within any model adopted by a financial institution to analyse its investment portfolio. The limitations to the application of stress testing are also discussed, as well as the minimum requirements set in this area by the Basle Committee on Banking Supervision.