Combining wavelet decomposition with machine learning to forecast gold returns

This paper combines the discrete wavelet transform with support vector regression for forecasting gold-price dynamics. The advantages of this approach are investigated using a relatively small set of economic and financial predictors. I measure model performance by differentiating between a statistically-motivated out-of-sample forecasting exercise and an economically-motivated trading strategy. Disentangling the predictors with respect to their time and frequency domains leads to improved forecasting performance. The results are robust compared to alternative forecasting approaches. My findings on the relative importances of such wavelet decompositions suggest that the influences of short-term and long-term trends are not stable over the full evaluation period.     

Open Economy Dynamics in a Floating Exchange Rate Developing Country Context

ABSTRACT

This article develops a wavelet-based control model to simulate fiscal, monetary, and real exchange rate scenarios in an open economy developing country with an inflation-targeting regime. We use South African macro data to jointly simulate optimal fiscal and monetary policy under varying scenarios for real exchange rate stability with interest rate parity. As real exchange rate stability increases, the model simulates the effects on the trade balance under both a constant and depreciating real exchange rate. We find that short-term cycle stability problems are somewhat mitigated by allowing the real exchange rate to depreciate.     

Time-frequency co-movements between oil prices and interest rates: Evidence from a wavelet-based approach

This paper investigates time–frequency co-movements between crude oil prices and interest rates. To test this relationship, the study applied a continuous wavelet and cross wavelet approaches to data from West Texas Intermediate (WTI) crude oil prices and interest rates in the United States (U.S.). Results from the sample period revealed significant relationships, in the intermediate term, between WTI crude oil prices and U.S. interest rates. Moreover, co-movements between oil price and interest rate variables were especially sensitive during abnormal political events and periods of financial ‘meltdown’. We further use Partial Wavelet Coherence (PWC) and Multiple Wavelet Coherence (MWC) methods to investigate the impacts of five major control variables namely GDP growth, unemployment, three-month Treasury bill, CPI index and industrial production index. The results show a powerful impact of control variables on oil-interest rates co-movements under different frequencies. Finally, we show evidence of co-integrating long run relationship between oil markets and control variables. These results have important implications for energy investors and policy makers.     

STOCK‐BOND CO‐MOVEMENTS AND FLIGHT‐TO‐QUALITY IN G7 COUNTRIES: A TIME‐FREQUENCY ANALYSIS

This paper examines co‐movement between stock returns and changes in 10‐year government bond yields as well as flight‐to‐quality behaviour in G7 countries. We conduct the wavelet squared coherence analysis to explore the dynamics in both time and frequency domain. Our results provide evidence of positive co‐movements, which vary over time and across investment horizon. The higher co‐movement is found to be more concentrated in the lower frequency bands. We further analyse the dynamic nature of the scale‐dependent wavelet correlations and find that the correlations are highly volatile and significantly increase across different time scales during the episodes of equity market turbulence. The increase in correlations reflects flights from stocks to safer bond investments as a result of dramatic changes in investor sentiment and risk aversion at times of market stress.     

利用双正交小波变换优化G3-PLC系统性能

为了提高G3-PLC(G3-Power Line Communication)在复杂信道环境下通信的可靠性，提出了基于双正交小波变换的正交频分复用（Discrete Wavelet Transform-Orthogonal Frequency Division Multiplexing,DWT-OFDM）。通过双正交小波变换减小复杂信道环境下G3-PLC通信系统中OFDM载波间干扰，并通过合理减少循环前缀，优化传输效率，降低峰均比。在DWT-OFDM的G3-PLC系统中，通过不同小波基性能分析与对比，利用bior6.8双正交小波基进行系统性能优化。将DWT-OFDM的G3-PLC系统在实际电力线噪声库环境下进行实验分析，仿真结果表明，在平均误码率为10-3量级时，性能有5 dB左右提升，通信效率由40%提高至54%；结合限幅法，峰均比由11 dB降低至8 dB，通信性能极大优于基于FFT-OFDM的G3-PLC系统。     

基于图形小波变换的遥感图像表示与去噪

离散小波变换（Discrete Wavelet Transform，DWT）通常用于图像的表示。然而，对于具有不规则形状边缘的图像，尤其是对于纹理和细节信息较多的遥感图像，DWT却很难有效表示，进而影响后续去噪效果。针对该问题，提出了一种基于图形小波变换（Graphic Wavelet Transform，GWT）的图像去噪方法。首先，将图像表示为图形信号，并通过该图形信号的谱表示构造相应的变换矩阵；然后，设计了一种改进自适应阈值的图像去噪方法，在GWT变换域内对图像去噪。实验结果表明，与常用的图像去噪方法相比，所提算法能够提供更好的图像主观质量。采用均方根误差（Root Mean Square Error，RMSE）和峰值信噪比（Peak Signal-to-Noise Ratio，PSNR）作为客观指标，结果表明，采用所提方法得到的重建图像客观质量更优。     

小波域抑制多音干扰的扩频接收机

利用小波包理论提出了一种快速、准确地识别并且抑制多音干扰的有效方法。该方法利用小波包理论将多音干扰的各条谱线逐个定位，然后通过功率谱分析的方法来识别多音干扰的存在区间，最后将被干扰污染的区间逐个切除并反向重构接收到的信号。实验证明该方法的误比特率性能在强干扰时要优于以前的FFT域切割法。     

基于小波变换的图像处理综述

小波分析主要广泛应用在科学研究和工程技术中。虽然在现阶段的小波理论相对成熟,近些年关于小波理论的应用和研究也在不断的发展和更新。小波变化在图像处理领域中的应用也囊括图像与处理的所有方面。本文通过介绍小波变换的起源,将小波应用在图像处理中的压缩、还原图像、边缘检测和图像分割,宏观剖析小波的研究现状历史、发展动向及优势。     

Does higher government debt link to higher social expenditure? New method,new evidence

Scholars believe that higher social expenditures are usually linked with higher government debts, whereas higher debts reduce social expenditures. However, it is reasonable to speculate that higher government debt may contribute to higher social spending, while fiscal deficits occur during a recession, which commonly creates greater demand for social expenditure. For a deeper investigation, this paper revisits the dynamic relationship between social spending and public debts in the time-frequency domain, using the novel wavelet-coherency analysis as well as the phase-difference technique to derive the co-moved and causal relationships between social spending and public debts in 13 OECD countries. The evidence identifies a dynamic relationship between variables. While higher social expenditures increase government debts, the shocks from government debts to social expenditures are conversely uncertain. We discover that higher government debt does reduce social expenditures, but it may be linked to higher social spending. The robustness of partial coherency and phase-difference discovers the role of a political party in the decision over social welfare programmes in the sample countries.     

基于小波变换和曲线拟合法的柴油机故障诊断

为了对柴油机故障类型进行准确诊断,对柴油机缸盖振动信号进行特征提取,所测取的信号通过小波变换提取时域、频域特征参数,得到相应的拟合曲线并计算多项式系数。设定两种常见故障,通过比较其小波变换拟合多项式系数与标准模式间的距离,从而判断出故障类型。经过验证,所判断出的故障类型正好为所设定的故障,由此表明该方法具有很大的优越性。