Skip to main content

Featured

关于对24年中国经济形势的一点看法

        今天已经是大年初五,春节也差不多接近尾声了,也是我在老家待的最后一天,刚好饭后闲来无事,终于静下心来有空写一写宏观经济分析。         回顾23年春节前的几个交易日,权益市场比较动荡,中证1000的平值隐含波动率最高冲到了91.48,要知道中证1000的实现波动率中位数也就15左右,而春节前几个交易日的连续大幅下跌和国家队快速出手使得权益市场走出深V形态,历史和隐含波动率也随之快速飙升。                另外伴随着雪球集体敲入、DMA爆仓等各类事件爆发,权益市场一片鬼哭狼嚎,就在大家都在讨论这波大A行情该谁来背锅时,证监会突发换帅。想想之前频繁出现在财经类流量博主文章中的北向、量化、公墓等,这次券商场外衍生品和私募微盘股应该也难逃一劫。都说经济繁荣时,大家都忙着数钱根本没有人在意合不合规,经济衰退时,你连呼吸都是错的,人性就是如此。关于现有微观市场体制的一些问题我之前也写过一些文章,这里不想再赘述,这里只想探讨一下宏观经济形势问题。         经济活动存在周期,这是我们初学经济学时就所熟知的,一个完整的经济周期包含繁荣、衰退、萧条和复苏四个阶段,每个阶段一般没有固定的时间长度和明显的分界线。但是如果回顾国内经济发展的历史情况,我们便可以大致发现国内经济增长开始下滑并不是近两年才开始的,三年疫情只是一场突如其来的黑天鹅,并没有影响整个大经济周期的演变方向。              从上图不难看出,从2001年加入世贸组织后,我国经济增长率同比逐年上升,呈现出快速发展的繁荣景象,也就是当时全球媒体称赞的“中国速度”。直到2008年,美国次贷危机爆发,中国也深受波及,随后政府出台了史上最大规模的“4万亿”扩张政策,虽然帮助中国摆脱了金融危机的泥潭,但也造成了后续非常严重的产能过剩、通货膨...

Total Pageviews

READING 24: MODELING AND FORECASTING TREND

Linear Trend Models
        A linear trend is a time series pattern that can be graphed with a straight line:
           yt = β0 + β1(t)
         A nonlinear trend is a time series pattern that can be graphed with a curve. Nonlinear trends can be modeled using either quadratic or exponential (i.e., log-linear) functions:
           yt = β0 + β1(t) + β2(t)2
           yt = β01(t) or ln(yt) = ln(β0) + β1(t)

        where:
           yt = the value of the time series (the dependent variable at time t)
           β0 = regression intercept at the vertical axis
           β1 = regression slope coefficient (or trend coefficient)
           t = time trend or time dummy (the independent variable): t = 1, 2, 3, … , T − 1, T
Estimating and Forecasting Trends
        Most statistical software packages can apply ordinary least squares (OLS) regression to estimate the coefficients in a trend line. The regression output can then be used to forecast in-sample and out-of-sample data.
Mean Squared Error
        Mean squared error (MSE) is a statistical measure computed as the sum of squared residuals (SSR) divided by the number of observations in a regression model:
The s2 Measure
        The unbiased MSE, s2, adjusts for the degrees of freedom, k, in the denominator as follows:

        The penalty factors for s2, Akaike information criterion (AIC), and Schwarz information criterion (SIC) are (T / T − k), e(2k / T), and T(k / T), respectively. SIC has the largest penalty factor.


 
Evaluating Consistency
        A selection criteria is considered to be consistent if the following two conditions are met:
  • When the true model or data generating process (DGP) is one of the defined regression models under consideration, then the probability of selecting the true model approaches one as the sample size increases.
  • When the true model is not one of the defined regression models being considered, then the probability of selecting the best approximation model approaches one as the sample size increases.
        The SIC is the most consistent selection criteria.

Popular Posts