Consider a time series transformed by an instantaneous power function of the Box-Cox type. For a wide range of fractional powers, this paper gives the relative bias in original metric forecasts due to use of the simple inverse retransformation when minimum mean squared error (conditional mean) forec
Loudness and the power series transformation
β Scribed by T. D. Clack
- Publisher
- Springer-Verlag
- Year
- 1978
- Tongue
- English
- Weight
- 953 KB
- Volume
- 28
- Category
- Article
- ISSN
- 0340-1200
No coin nor oath required. For personal study only.
β¦ Synopsis
A nonlinear-algebraic approach to monaural intensity processing is proposed: the purpose is to integrate the conscious loudness attribute evoked by a pure tone with the power-series description of auditory distortion. The preliminary model consists of six postulates leading to a mathematical definition for loudness. To evaluate this equation, two task-specific sets of loudness judgments from each of seven subjects are examined. When linked to loudness-interval responses via the equisection assumption, the equation describes the behaviors quite well. Extrapolations into other intensity ranges predict similar responses with relatively slight overestimates. By relaxing the assumption that subjects will adjust the loudness proportions exactly as instructed, their ratio productions also can be described and predicted with sometimes surprising accuracy. Particularly striking is the prediction of overall levels and the curvilinearities of "doublings" from the "halvings". In addition, the theory proposes absolute loudness measurement, an explanation for the growth of loudness including the principle underlying Steven's Power Law, and might prove useful in examining some exceptions to this relationship. Several aspects of this model differ from traditional approaches to intensity processing, but it appears to warrent further critical evaluations.
π SIMILAR VOLUMES
A simple recursion is presented for calculating moments (e.g., mean, variance, and autocorrelation function) of a time series that has been power transformed to normality. Its derivation is elementary, relying on the moment-generating function for a bivariate normal distribution. To make clear the d
**RUNAWAYS. ROMANCE. ROAD TRIPS.** Sheβs guarded and angry at humanity. Heβs a sensitive guitar player who is trying to escape his family drama. They donβt seem like theyβd be suited for one another, but it turns out they may be the perfect match to get each other through their complicated obstacle
In this paper we discuss procedures for overcoming some of the problems involved in fitting autoregressive integrated moving average forecasting models to time series data, when the possibility of incorporating an instantaneous power transformation of the data into the analysis is contemplated. The
## Abstract This paper presents some procedures aimed at helping an applied timeβseries analyst in the use of power transformations. Two methods are proposed for selecting a varianceβstabilizing transformation and another for biasβreduction of the forecast in the original scale. Since these methods