A comparison of state space and multiple regression for monthly forecasts: U.S. Fuel consumption

Monthly consumption forecasts for U.S. oil, natural gas, and coal are made using state space and multiple regression applied to the same data. These forecasts are compared with actual consumption for a test period. The forecasts made using state space are preferred to those made using multiple regression models for both expost and exante cases. The state space forecasts track data cycles better than do the regression forecasts. Average absolute forecast errors are less for the state space models than they are for the multiple regression models.     

船舶气象导航随船验证

通过横流太平洋往返两个航次随船调查，研究了采用气象导航对减少燃油消耗和保证船舶安全、班期等，不但有定的经济效益，而且是一种非常有效的航运管理方法。     

1823 年(清道光三年)我国特大水灾及影响

该文对道光三年（1823年）长江下游与华北的严重洪水概况及其对社会经济的影响作了初步探讨，并对这次洪水发生的原因及灾害形成的机制也作了分析，认为了解这次洪水及其产生的影响对于我国减灾战略的制订有重要参考价值。     

用时间序列模拟和预报天山乌鲁木齐河源1号冰川年径流总量

本文把1958—1984年天山乌鲁木齐河源1号冰川的融水径流总量的时间序列X(t),分解为趋势项L(t)、周期项P(t)、平稳项S(t)、随机项ε(t),使该时间序列表示为X(t)=L(t)+P(t)+S(t)+ε(t)。采用非线性回归提取L(t),用谱分析和Fourier级数提取P(t),余差用自回归方程建模,用上述项的叠加作出预报,按相关指数公式计算R=0.90,效果良好。     

Observations of desert flood bores

Flood bores have been measured in desert stream floods. The bores were steep and small pebbles were observed to be pushed ahead. Bore velocity changed downstream and was controlled by local channel geometry. In narrow reaches, the bore advanced at rates twice those of wide reaches. Mean bore velocity was about 50 per cent of that of mean flow at peak flood discharge. The surfaces of shallow bores were covered by air foams. This was not the case in deeper, faster examples.     

A univariate model for long-term streamflow forecasting

This paper, the first in a series of two, employs the principle of maximum entropy (POME) via maximum entropy spectral analysis (MESA) to develop a univariate model for long-term streamflow forecasting. Three cases of streamflow forecasting are investigated: forward forecasting, backward forecasting (or reconstruction) and intermittent forecasting (or filling in missing records). Application of the model is discussed in the second paper.     

Entropy estimation of hydrological extremes

The method of Relative Entropy with Fractile constraints (REF method) is explained and applied to model extreme compound hydrological phenomena, such as extreme sea levels under storm conditions. Also presented is a simple method of Tail Entropy Approximation (TEA), which amounts to a correction of traditional statistical estimates for extreme observations.Distribution assumptions are necessary but downplayed in the REF method, relegating the prior distribution to the role of an extrapolation function. The estimates are objective in an information-theoretical sense. They also satisfy a strict requirement of self-consistency that is generally not satisfied by standard statistical methods: invariance under monotonic transformations of the random variable.Historical records of storm surge levels in the Netherlands and annual maximum tidal heights for Sheerness, UK, are used as examples. Comparison is made with distributions obtained using other methods.It is concluded that the tail entropy approximation provides simple, objective estimates of extremes in the tail beyond the range of observations.     

PRELIMINARY STUDY OF BAROTROPIC STOCHASTIC DIFFERENCE MODEL APPLIED TO WEATHER PREDICTION

In order to consider both the deterministic and the stochastic property of atmospheric motion simul-taneously,in this paper,the weather prediction is proposed as the problem of the evolution of meteorologicalfield.The historical viewpoint of atmospheric motion is emphasized here.Based on time series analysis te-chnique,a stochastic-dynamical model with multiple initial fields is derived.Thus,weather forecasting is sum-meal up as a problem of solving a set of stochastic difference equations.For the barotropic atmosphere,thenumerical solutions of the equations are obtained by using the method of empirical orthogonal functions(EOF),and examples of medium-range weather prediction are given here.Meanwhile,selecting the order oftime series,i.e.,determining the number of initial fields properly,is also discussed.