理解exponential weighted || 指数

• Exponentially weighted moving average

  average is a single data point;

  moving average is a sequence of data point;

  weighted moving average means the process calculating average 给不同元素不同权重;

  exponential weighted moving average 指计算权重的方式是指数方式;

  先看这四个概念中最核心的average，求average的目的在物理意义上是smoothing the data: reduce the noise in noisy time-series data.

• EWMA formula

  $$S_t=a{x}_t+(1-a)S_{t-1}$$

  最新值给予权重$a$;

  前值给予权重$1-a$;

  指数体现在前值中。

  The degree of mixing is controlled by the parameter $a$ (value between 0-1).

  If a=10% (small), most of the contribution will come from the previouse value of the signal. In this case, “smoothing” will be very strong.

  If a=90%(large), most of the contribution will come from the current value of the signal. In this case, “smoothing” will be minimum.

• Exponential

  If $n$ is a positive integer and $x$ is any real number, then $x^n$ corresponds to repeated mutiplication
  $$x^n=x\ast x\ast ...\ast x$$
  We can call this “x raised to the power of n”;

  or “x to the power of n”;

  or “x to the n”;

  Here , x is the base and n is the exponent or the power.

  An exponential function can describe growth or decay.
  $$g(x)=(\frac{1}{2}{)}^x$$
  is an example of exponential decay.
  $$f(x)=2^x$$
  is an example of exponential growth.

  In the exponential growth of $f(x)$ , the function doubles every time you add one to its input $x$;

  In the exponential decay of $g(x)$, the function shrinks in half every time you add one to its input $x$.

  The presence of this doubling time or half-life is characteristic of exponential functions, indicating how fast they grow or decay.

  指数方程用于描述增长或衰减的速度。

• References

1. [Beginner]ML Basics: Exponentially weighted moving average
2. Math Insight : The exponential function