Probability distributions,trading strategies and leverage: an application of Gaussian mixture models

The purpose of this paper is twofold. Firstly, to assess the merit of estimating probability density functions rather than level or classification estimations on a one‐day‐ahead forecasting task of the EUR/USD time series. This is implemented using a Gaussian mixture model neural network, benchmarking the results against standard forecasting models, namely a naïve model, a moving average convergence divergence technical model (MACD), an autoregressive moving average model (ARMA), a logistic regression model (LOGIT) and a multi‐layer perceptron network (MLP). Secondly, to examine the possibilities of improving the trading performance of those models with confirmation filters and leverage. While the benchmark models perform best without confirmation filters and leverage, the Gaussian mixture model outperforms all of the benchmarks when taking advantage of the possibilities offered by a combination of more sophisticated trading strategies and leverage. This might be due to the ability of the Gaussian mixture model to identify successfully trades with a high Sharpe ratio. Copyright © 2004 John Wiley & Sons, Ltd.     

一类图的匹配唯一性

本文证明了图κK1∪m2P2∪m3P3∪[∪↑i≥2m2ipi]∪dD4∪tT1、2、3∪sT1、2、4匹配唯一当且仅当dm2=dm3=0，其中κ、m2、m3、m2i(i≥2)、d、t、s都是非负整数。     

沥青生产过程的线性回归控制

本文利用线性回归的方法给出沥青软化点与电阻率和温度的函数关系式。     

基于提升静态小波变换的自适应消噪方法

基于离散小波变换的自适应消噪方法为雷达信号的滤波提供了一种可行的方法.但DWT不具有平移不变性,若不用相同的小波对滤波后的信号进行重构,则会带来较大的重构误差.针对这一现象,提出了基于提升静态小波变换的自适应消噪方法,它首先根据DWT的提升方法,得到SWT的提升和对偶提升实现方法,然后通过SWT的提升方法将信号分解为多个子带,利用引入更多动量因子的权系数的迭代公式进行自适应匹配.并对匹配结果二次自适应,得到拟合的原信号.仿真结果表明,该方法可在计算量增加不大的前提下,进一步改善系统的滤波性能.     

Function approximation for incompletely specified regression models

This report extends earlier work by Brailovsky on regression theory and methodology, giving particular emphasis to function approximation for incompletely specified models. The interest here is with situations where the form of the regression relation is not known in advance. We discuss several difficulties that arise in using local approximation and linear regression methods, and propose ways to overcome these problems. To aid the data analyst in developing a suitable model, an illustrative table is derived for determining the number of initial explanatory functions justifiable for a given prespecified confidence level. The general approach formulated here is illustrated with an application to medical data. Relevance to classification and possible extensions are discussed.     

阿氏公式参数估计方法的分析与改进

对于阿维里扬诺夫潜水蒸发公式的参数估计问题,本文提出两种新的解法,即模拟相关分析法和解析法,经与传统的作图相关法相比较,表明两种新方法都是可行的,尤以模拟相关分析法最为可靠和方便。     

STRONG CONSISTENCY OF M-ESTIMATES OF MULTIPLE REGRESSION COEFFICIENTS

ＳＴＲＯＮＧＣＯＮＳＩＳＴＥＮＣＹＯＦＭ－ＥＳＴＩＭＡＴＥＳＯＦＭＵＬＴＩＰＬＥＲＥＧＲＥＳＳＩＯＮＣＯＥＦＦＩＣＩＥＮＴＳＣＨＥＮＸｉｒｕ（ＧｒａｄｕａｔｅＳｃｈｏｏｌｏｆｔｈｅＣｈｉｎｅｓｅＡｃａｄｅｍｙｏｆＳｃｉｅｎｃｅｓ，Ｂｅｉｊｉｎｇ１００...     

关于带有传输时间的F2|perm|Cmax排序问题的一个新算法

考虑传输时间的同顺序排序问题是在机器人设计及柔性制造系统中有着广泛应用的一类排序问题。     

一类图的色等价问题

本文给出了某类图色多项式的一个计算公式，可以部分地改进文［１］中ｋ４同胚图象色项式的推导过程，并定义了图的ｋ－包和紧ｋ－包。通过上述公式计算得出ｋ－包与紧ｋ－包的色多项式与原图色多项式的关系。     

Rm的置换多项式与COSTAS矩阵

本文给出了多项式ｈｋ（Ｘ）＝１十ｘ十ｘ２＋…＋ｘｋ置换剩余类环Ｒｍ（Ｚ／（ｍ））的充分必要条件；证明了Ｒｍ的纯奇次置换多项式ｆ（ｘ）不能生成ｍ×ｍ（ｍ≥４）的Ｃｏｓｔａｓ矩阵．并深人讨论了迪克逊多项式ｇｋ（ｘ，ａ）和ｈｋ（ｘ）与Ｃｏｓｔａｓ矩阵的关系．