Forecasting the Yen/U.S. Dollar exchange rate: Empirical evidence from a capital enhanced relative PPP-based model

This study uses a relative purchasing power parity (PPP) model based on price indexes (consumer, CPI or traded-goods price indexes, TPI), interest rate differentials, and a linear forecasting technique to determine the horizon over which such a model outperforms a random walk in forecasting the Yen/U.S. Dollar exchange rates out-of-sample. The results improve if one adjusts a simple CPI-based PPP-model by interest rate differentials, while the best results are obtained using a TPI-based PPP-model. For example, the TPI-based model, adjusted by interest rate differentials, is able to statistically significantly outperform the pure random walk starting at forecast horizons of 1 month.     

The rationale for projected prices of demand on electricity for industry in Russia up to 2020

The paper presents a forecast of the demand price for commodities, which can be considered dual to the forecast for supply price. An algorithm for preparing the initial information is presented that allows one to take into account the features of the original data that determine the quality of the evaluation of the model parameters. We describe the extraction of input data for predicting the demand price from the data intended for determining the supply price.     

基于因果聚类分析理论的维修成本控制体系研究

供电网络的维修维护工程比较复杂,影响的因素较多,影响方式多种多样,这给成本控制造成了很大困难。为了解决这个问题,我们采用数据分析,建立预测模型的方法,通过分析预测误差和预测的置信区间,确定维护成本的控制范围。为了提高预测得准确性,利用历史数据对要素变量进行因果聚类。然后将各类型中的变量进行主成份分析,降低变量集的维数。对选出的变量用神经网络算法构建预测模型,并利用其误差的性质确定误差范围,实现对成本的控制。     

A medium-term forecast of crude oil buyers’ prices

This paper describes the forecasting of crude oil prices for the next years. This forecast is essential to strategic investors' understanding of the price at which it will be possible to sell the goods in which they have invested.     

A STOCHASTIC ESTIMATION FRAMEWORK FOR COMPONENTS OF THE SOUTH AFRICAN CONSUMER PRICE INDEX

Under inflation targeting in South Africa, it is important to monitor and forecast changes in prices, not only for aggregate measures of the consumer price index, but also its underlying sub-components. Hypotheses about sectoral transmission of policy and shocks are often more specific than hypotheses about overall transmission. This study employs a stochastic framework to estimate richly specified equilibrium correction models, four-quarters-ahead, for the 10 sub-components of the first targeted measure of the consumer price index, CPIX. The stochastic trends are estimated by the Kalman filter, and interpreted as capturing structural breaks and institutional change, a frequent cause of forecast failure. The trends suggest the design of deterministic split trends for use in recursive forecasting models, towards more accurate overall inflation forecasting. This research also has practical use for monetary policy in allowing identification of sectoral sources of inflation.     

联合谐波小波与递归神经网络的股市时间序列预测

为了提高神经网络对股市时间序列的预测精度，首先利用谐波小波对股票市场数据进行多尺度分解，将其分解为不同尺度且具有平移不变特征的谐波小波分量；然后根据股市时间序列的特点，构建递归神经网络模型进行短期预测，以不同尺度的谐波小波分量为输入数据，对股市数据进行多尺度预测；最后对不同尺度的预测结果进行谐波小波重构，得到最终的股市预测数据。对我国股票市场进行了实验分析，结果表明：股市时间序列经谐波小波分解后，股市数据中不同投资时间水平的价格波动可以被较好的分离，有效地提高了股票市场数据的预测精度。     

The impact of world oil price shocks on macroeconomic variables in Vietnam: the transmission through domestic oil price

This study examines the impact of world oil price shocks on macroeconomic variables in Vietnam with a focus on the transmission channel of domestic oil prices. The Structural Vector Autoregression model with two blocks of real economy variables and monetary variables is employed. The world oil price follows an autoregressive process to reflect the exogenous nature of world oil price shocks to the domestic economy. The retail domestic oil price is determined simultaneously by only the world oil price due to the government's control of the domestic oil market. Using monthly data in the period between 2009 and 2021, the study indicates that a positive shock to world oil prices will increase the domestic oil prices significantly, industrial production (slightly and only statistically significant in the third month after), and inflation (significantly in 8 months). Besides, the domestic oil price is not the only transmission channel of world oil price shocks to the economy. This result implies forecasting, assessing, and controlling the impact of the world oil price shock on the economy should focus on both domestic oil prices and other indirect channels.     

Do professional forecasters apply the Phillips curve and Okun's law? Evidence from six Asian-Pacific countries

Using survey data from six Asian-Pacific countries, we report that professional forecasters apply the wage Phillips curve, the price Phillips curve, and Okun's law when forecasting macroeconomic variables. This result is robust when using time-varying coefficients, different forecast horizons and when taking business-cycle asymmetries into account. The results also suggest that the confidence in macroeconomic relationships was more pronounced during the economic crisis 2007–2009 and when looking at longer forecast horizons.     

FORECASTING SOUTH AFRICAN INFLATION

This paper looks at what variables are useful for forecasting inflation starting in 1990. I show that the output gap, a measure of real economic conditions, does seem to provide useful information for forecasting inflation. This is good news for the Reserve Bank, since the primary way that the Reserve Bank tries to affect future inflation is through real economic conditions. In addition, short‐term interest rates and import price inflation also seem to provide useful information. The most accurate of these forecasts suggests a root mean square forecast error of 1–2 per cent for 1‐year ahead inflation, which is within the Reserve Bank's current target range.     

一种对指数增长年度用电量的组合预测方法

为有效预测呈指数增长的年度用电量,本文选用了指数函数法、GM(1,1)灰色模型和多元线性回归模型来对年度用电量进行拟合外推预测。通过运用这些方法预测北京年度用电量和对预测结果进行比较分析得到,GM(1,1)灰色模型的实际预测结果往往偏大,指数函数法预测结果可能偏小,而多元线性回归模型虽然考虑了其他因素影响但在实际预测指数增长负荷时参考意义较小。因此,最终选择合适的权重加权组合以上模型得到最终预测值,预测精度较高。     

Interrelations between the dynamics of oil prices and demand: Contemporary characteristics

The article shows that the unprecedented rise in oil prices was in recent years accompanied by an increase in oil consumption. The author considers the growth that was seen in the magnitude of oil sales despite multiple increases in price to be, along with the buying up of oil as a high liquidity asset, a phenomenon that indicates an approaching crisis and allows us to create new approaches to its forecasting.     

企业需求预测准确性存在的问题及其应对策略

需求预测在企业经营活动中具有重要作用,其影响因素具有多样性.可以通过优化需求预测方法、完善需求预测系统、改进需求预测管理,有效控制需求预测和未来市场情况的偏差,持续提高需求预测的准确性,从而促进企业生产、销售的良性运行.     

Dynamic forecasting of sticky-price monetary exchange rate model

The Dornbusch-Frankel monetary model is used to estimate the out-of-sample forecasting performance for the U.S. or Canadian dollar exchange rate. By using Johansen's multivariate cointegration, up to three cointegrating vectors were found between the exchange rate and macroeconomic fundamentals. This means that there is a long-run relationship between exchange rate and economic fundamentals. Based on error-correction models, the random-walk model outperforms the Dornbusch-Frankel model at every forecasting horizon. The random-walk model also dominates the Dornbusch-Frankel model with the modified money demand function at every forecasting horizon except one month. However, this paper shows that the share price variable can improve the accuracy of forecasts of exchange rates at short-run horizons.     

铁路货运量预测过程中的关键技术分析

可靠的铁路货运量预测对铁路部门计划、建设、运营等决策具有重要意义。主要影响因素筛选、预测方法选取等是预测过程中的关键技术问题,对预测质量有重要影响。基于已有研究成果,首先分析了预测机理,探讨了影响因素的界定方法;接着提出了主要影响因素的识别原则和策略;然后归纳出各主要预测方法的优缺点;最后给出相关建议,以期更加科学合理地预测铁路货运量。     

Forecasts for international financial series with VMD algorithms

Recent models with variational mode decomposition (VMD) have been applied to time-series forecasting. In this paper, we build a hybrid model named VMD–autoregressive integrated moving average (ARIMA)–Taylor expansion forecasting (TEF) to increase accuracy and stability for predicting financial time series. We use VMD algorithms to decompose financial series into subseries. An ARIMA model is built to predict each mode’s linear component, and the pragmatic TEF model based on a tracking differentiator is applied to forecast of the nonlinear component. Then the forecasts of all subseries are assembled as a final forecast. Our empirical results of international stock indices demonstrate that the proposed hybrid approach surpasses several existing state-of-the-art hybrid models.     

Forecast of global steel prices

The article presents a regression model for forecasting global average annual steel prices. As factors of price changes, we used the coefficient of load capacity of world steel plants and prices for iron ore and coking coal. On the basis of these factors, steel prices were predicted for 2010–2012.     

Evaluation of variants of economic regulation with the help of interregional “payments-income” tools

A pioneering experience in upgrading interregional models of intersectoral balance to “payments-income” systems that involve the financial side of production ties is described. Optimizing models of this class have been combined with tools for calculating aggregative price indices, which allows one to make up summary intersectoral, regional, and interregional financial balances for different forecast variants of the physical (material) production pattern.     

The implications of monetary expansion in China for the US dollar

We examine the influence of rapid growth in China's money supply on the US dollar within a framework of monetary models of exchange rates. We develop out-of-sample forecasts of the US dollar exchange rate using US and global data on price level, output, and interest rates, and money supply data for the US, China, and the rest of the world for the period 1996–2013. Monetary model forecasts significantly outperform a random walk forecast in terms of mean squared forecast error in the long run. A monetary error correction model with sticky prices performs best. Rolling sample analysis indicates changes over time in the influence of Chinese money supply in forecasting the US dollar. The expectation is that rapid money growth in China would increase the demand for dollars thus raising the value of the dollar, yet our forecasts are to the contrary for the mid 2000s. This is consistent with anticipation of renminbi appreciation under China’s managed exchange rate, which made holding renminbi more attractive. With the break from a dollar peg in 2005 and subsequent currency appreciation, the distortion was alleviated and the forecast direction for the dollar became as expected.     

基于组合模型的股票预测研究综述

随着金融业数据环境的日益复杂,利用传统单一模型进行高精度股价预测变得愈加困难。面对日益突出的股票分析技术需求,组合预测模型开始得到发展并取得了很多成果。首先介绍影响股价波动的分析指标,概括基于传统统计预测模型、机器学习、神经网络等单一预测模型在股票预测中的优势与不足。然后依据组合预测模型的组合形式,将其分为线性模型的组合、非线性模型的组合以及线性与非线性模型的组合3种类型,并报告多种组合模型的实际应用与研究现状。最后,对组合模型股票预测方法的有效性和稳定性进行展望。     

基于EMD-SE-LSTM模型的股指日内已实现波动率预测──以中证500指数为例

由于股指波动率具有非平稳、高嘈杂、非线性等特征,而传统的预测模型在建模时要求数据平稳、线性或近似线性,所以很难精准预测股指波动率。为提高股指波动率的预测效果,采用经验模态分解(EMD)、样本熵(SE)和长短期记忆网络(LSTM)构建的模型对股指日内已实现波动率进行预测。以中证500指数为例,经过EMD分解得到一系列分量,再根据分量的样本熵大小进行重构,最后利用LSTM对重构后的各序列进行预测。结果表明,EMD算法对LSTM模型的预测精度有很大的提升,相较于传统模型,EMD-SE-LSTM模型在预测股指波动率时精度更高,拟合优度更好。