C63 - Computational Techniques; Simulation ModelingReturn

Random forest models have recently gained popularity for economic forecasting. Earlier studies demonstrated their potential to provide early warnings of recession and serve as a competitive method to older prediction models. This study offers the first evaluation of the random forest forecast for the Czech economy. The one-step-ahead forecasting results show high accuracy on the Czech data and are proven to outperform forecasts from the Czech Ministry of Finance and the Czech National Bank. The following multi-step random forest forecast, estimated for the next four quarters, shows results similar to those from the central institutions. The main difference stems from the household and industrial confidence variables, which significantly impact on the random forest forecast. The variable-importance analysis further emphasizes the soft variables as valuable determinants for Czech forecasting. Overall, the findings motivate other forecasters to exercise this method.

One of the major limitations of the supply-driven input-output (I-O) Ghosh model concerns its linear production function. Using the I-O table, this paper replaces the linear production function with the Cobb-Douglas (CD) production function within the supply-driven model. The two models are compared both theoretically and empirically. Nonlinear production function, relative substitutability of primary factors, and variability of the proportion of intermediate inputs over product levels are the characteristics of the proposed model. The consideration of sectors' Solow residual as Total Factor Productivity (TFP) of sectors is yet another characteristics of the proposed model. The model is also plausible in value added and supply shock computations.

The aim of this paper is to shed light on Collateralized Debt Obligation (CDO) valuation based on data before and during the 2007-2009 global turmoil. We present the One Factor Gaussian Copula Model and examine five hypotheses regarding CDO sensitivity to entry parameters. For our modelling we used data of the CDX NA IG 5Y V3 index from 20 September 2007 until 27 February 2009 and we appropriately transform its quotes into CDO quotes. Based on the results we discovered four main deficiencies of the CDO market: i) an insufficient analysis of underlying assets by both investors and rating agencies; ii) investment decisions arise from the valuation model based on expected cash flows, they neglected other factors such as mark-tomarket losses; iii) mispriced correlation; and finally iv) obligation of the mark-to-market valuation. Based on the mentioned recommendations we conclude that the CDO market has a chance to be regenerated but in smaller volumes compared to the pre-crisis period. However, it would then be more conscious, driven by smarter motives rather than by pure arbitrage and profit incentives.