Abstract

Mixed-frequency variables may encounter problems of non-guaranteed steady-state in the time-variant state-space system during temporal disaggregation, forecasting, or nowcasting. The instability of the state-space system directly affects the accuracy of prediction. This paper aims to develop a new design framework to model non-stationary mixed-frequency variables in time series models. We introduce a periodic constraint to control the instability of the time-variant state-space system for non-stationary mixed-frequency variables. Our proposed periodic constraints in a time-variant state-space system originate from temporal-aggregated constraints themselves. Fully utilizing the binding conditions of both unobserved and observed temporal-aggregated conditions can generate the bounded periodicity of Kalman gain, control the instability of the time-variant state-space system and improve the accuracy of temporal prediction. Mixed-frequency variables in a periodic constrained state-space system are implementable with a simple conventional Kalman filter.

Part I

• https://rpubs.com/HeatWave2019/784639

Part II

Summary and Conclusion

Bibliography

Appendix