1 / 18

SpaeSoft Managerial risk assessment services

SpaeSoft Managerial risk assessment services. Software. SpaeSoft delivers high precision software for strategic firm planning and risk assessment, investment analysis, forecasting, classification and optimization problems in corporate expert systems, decision support systems,

hume
Download Presentation

SpaeSoft Managerial risk assessment services

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. SpaeSoftManagerial risk assessment services

  2. Software SpaeSoft delivers high precision software for • strategic firm planning and risk assessment, • investment analysis, • forecasting, • classification and • optimization problems in • corporate expert systems, • decision support systems, • logistics, • sales management and • production scheduling systems.

  3. The methodological basis • SpaeSoft applies techniques of Artificial Intelligence (AI) combined with classical optimization methodology • Computationally demanding simulation tasks (e.g., lead-lag orders in forecasting systems, structural decisions concerning neural networks, etc) may be automatized using AI.

  4. Target systems in risk assessment • Risk assessment of the firm as a multiperiod decision system • Measuring the risk of individual investments • Forecasting performance of financial instruments and portfolio optimization • Classifying objects in nonoverlapping groups (e.g., quality control in industrial processes: working/failing products; banking: healthy/financially distressed firms) -> Possibilities to significant cost reductions through, e.g. improved production scheduling, management of logistics and explicit recognition of risk in investment decisions.

  5. Multiperiod firm planning • The financial statements, firm decisions and parameters for a planning horizon form an intertwined unity, a scenario: where xt = the key decisions of the firm in period (year) t (sales, production, R & D, investments, financing, etc), pt = all relevant parameters controlling the operating environment of the firm (prices, costs, interest rates, exhange rates, chocks, economic regulation, legislation, etc), F = the vector function generating all financial statements and financial ratios according to strict accounting logic, H = planning horizon

  6. Multiperiod firm planning cont. The anatomy of the scenario P/Lt=profit and loss statement for year t BSt=balance sheet CFt=cash flow statement FRt=financial ratios

  7. Multiperiod firm planning cont. • Assume that management has specified a strategy using human reasoning, high performance optimization technology or a combination of both, for example over the period 2008-2012: • We want to assess the risk surface pertaining to this strategy when the unknown parameters (pt) are stochastic. • The probability distributions are assumed to be mutually independent, multivariate dependence is recognized within the distributions, i.e. within subsets of the parameters. • The distributions are extracted by statistical methods combined with human reasoning. • With pt stochastic, the scenarios as a whole will be stochastic as well:

  8. Multiperiod firm planning cont. • Risk assessment is carried out through Monte Carlo Analysis, a well-known and robust simulation technique. • A large number (M) of drawings of the uncertain ptare made from its underlying univariate/multivariate distributions • wages, sales prices, R & D, investment costs, raw material costs, interest rates, exchange rates, delivery times, manufacturing times, credit risk, machine failure, etc. • The scenarios will be approximately normally distributed by the central limit theorem under fairly general conditions. • The distribution of the simulated scenarios forms the estimated risk surface of the planned decisions.

  9. M random drawings of the uncertain pt from the underlying distributions and the resulting scenarios: Multiperiod firm planning cont.

  10. Multiperiod firm planning cont. A suitable financial criterion, for example, a critical minimum level on the return on investment – ROIc (or EBIT, Cash Flows, discounted Net Profit, etc) - may be used as a criterion for accepting/rejecting the strategy:

  11. Multiperiod firm planning cont. • The quantitative risk profile can be obtained • for any item or all items simultaneously in the financial statements • over any planning horizon • The method is connectable to the technological/economic processes of the firm • R & D, investments, logistics, production processes, pricing decisions, marketing campaigns, etc.

  12. Multiperiod firm planning: a summary • Input: • The managerial strategy based on • Human reasoning • Multiperiod optimization • Accounting logic: • Financial statements, • Uncertain parameters (prices, wages, interest rates, etc) • Monte Carlo Analysis: • Probability distributions • (statistical/judgemental) • Large-scale simulations Output: The risk profile of all key financial items

  13. Risk and project appraisal • In this case the scenario represents the investment project. • xtcontains the profile of the planned investment, e.g. xt(i) = the number of units of investment object i acquired at time point t. • pt = {ct+, ct-, rt} contains all relevant parameters, e.g. ct+(i) & ct-(i) = cash inflows & outflows per unit of investment object i at time point t, {rt(j), rt(k), rt(m)} = {the interest rate, the cost of equity, the weighted cost of capital (WACC)} at time point t, etc. • Fis the evaluation function for the investment, in the simplest case the net present value formula (NPV).

  14. Risk and project appraisal • The probability of the project falling below the critical level NPVc is obtained by Monte Carlo simulation.

  15. Monte Carlo Analysis • Some powerful results concerning the central limit theorem extend it to cases with finite variance and non-identical distributions as well as non-independent random numbers (Lindeberg [1922], Doukhan et al [1994]). • The method is used extensively in engineering, statistics, numerical mathematics and computational finance (e.g., Boyle, P (1977): A Monte Carlo Approach, Journal of Financial Economics, Hull J,C (2000): Options, futures and other derivatives (4th ed.)). • It is used scarcely in the Scandinavian accounting profession eventhough it has an obvious potential in risk assessment for financial planning and investment problems.

  16. The computational platform of SpaeSoft: The Genetic Hybrid Algorithm (GHA) • Spaesoft has its own platform for mathematical analysis, GHA, written in strict ANSI C and Fortran 90 • Useful for High-Performance computing on massively parallel supercomputers as well as on individual work stations/main frame computers

  17. The computer environment • SpaeSoft modules can be integrated into the corporate software. • Availability: UNIX, Alpha & LINUX mainframe computers and massively parallel IBM and Cray supercomputers. • SpaeSoft provides consultancy to the customer in model building and data gathering within the organization.

  18. References • Doukhan P, Massart P, Rio E (1994): The functional central limit theorem for strongly mixing processes. Annales de l’I.H.P., section B, tome 30, no 1, 63-82 • Lindeberg J.W (1922): Eine neue Herleitung des Exponentialgesetzes in der Wahrscheinlichkeitsrechnung. Mathematische Zeitschrift 15, 211-225. • Östermark, R. and K. Söderlund (1999): A multiperiod firm model for strategic decision support, Kybernetes28:5, pp. 538-556. • Östermark, R., H. Skrifvars, and T. Westerlund (2000): A nonlinear mixed integer multiperiod firm model, International Journal of Production Economics67, pp. 183-199. • Östermark, R. (2003): A multipurpose parallel Genetic Hybrid Algorithm for nonlinear nonconvex programming problems. European Journal of Operational Research,152, 195-214. • Östermark, R., J. Aaltonen, H. Saxén, and K. Söderlund (2004): Nonlinear modelling of the Finnish banking and finance branch index, European Journal of Finance10, pp. 277-289. • Östermark, R. (2007): A flexible platform for mixed-integer non-linear programming problems. Kybernetes. The International Journal of Systems and Cybernetics.36, No 5/6, pp. 652-670.

More Related