Predicting turning points in business cycles by detection of slope changes in the leading composite index

Article Abstract:

A Bayesian technique for detecting changing random slopes, or turning points, in leading composite indexes is proposed. Predicting the turning points of business cycles is one of the most essential forecasting problems for decision makers and economists. The underlying process of the composite leading index is detailed by a dynamic linear model with random leveland slope, where a major slope change at every turning point is represented by a random shock.

Methods, Bayesian statistical decision theory, Bayesian analysis, Business cycles

---

State space trend-cycle decomposition of the ARIMA (1,1,1) process

Article Abstract:

A trend-cycle decomposition of autoregressive integrated moving average (ARIMA)(1,1,1) process can undertaken using a state space (SS) model. SS trend-cycle decomposition has the tendency to become spurious under an unobservable estimated SS model. Redundant autoregressive (AR) and moving average (MA) parameters of ARIMA process result whenever an independence assumption is made between the trend and cycle's noise processes.

Econometrics & Model Building, Econometrics, Business models

---

Estimation and forecasting of long-memory processes with missing values

Article Abstract:

A method incorporating state-space models with the Kalman filter can be used to assess the possibility of an efficient estimation and forecasting of a long-memory time series with missing values. The method permits a productive estimation of a fractionally integrated autoregressive moving average (ARIFMA) process and future values with missing data. Its performance was proven effective using a foreign exchange data set.

Kalman filtering

Similar abstracts:

• Abstracts: Efficient Bayesian inference for multiple change-point and mixture innovation models. Testing for neglected nonlinearity in long-memory models
• Abstracts: Nonshiftable capital, affine price expectations and convergence to the Golden Rule
• Abstracts: Bootstrap methods for median regression models. Identification and robustness with contaminated and corrupted data
• Abstracts: The transition from a Dreze equilibrium to a Walresian equilibrium. Equilibrium adjustment of disequilibrium prices
• Abstracts: The method of simulated scores for the estimation of LDV models. An information-theoretic alternative to generalized method of moments estimation

This website is not affiliated with document authors or copyright owners. This page is provided for informational purposes only. Unintentional errors are possible.