Kernel Density Estimation - Relation To The Characteristic Function Density Estimator

Relation To The Characteristic Function Density Estimator

Given the sample (x1, x2, …, xn), it is natural to estimate the characteristic function φ(t) = E as

\hat\varphi(t) = \frac{1}{n} \sum_{j=1}^n e^{itx_j}

Knowing the characteristic function it is possible to find the corresponding probability density function through the inverse Fourier transform formula. One difficulty with applying this inversion formula is that it leads to a diverging integral since the estimate is unreliable for large t’s. To circumvent this problem, the estimator is multiplied by a damping function ψh(t) = ψ(ht), which is equal to 1 at the origin, and then falls to 0 at infinity. The “bandwidth parameter” h controls how fast we try to dampen the function . In particular when h is small, then ψh(t) will be approximately one for a large range of t’s, which means that remains practically unaltered in the most important region oft’s.

The most common choice for function ψ is either the uniform function ψ(t) = 1{−1 ≤ t ≤ 1}, which effectively means truncating the interval of integration in the inversion formula to, or the gaussian function ψ(t) = e−π t2. Once the function ψ has been chosen, the inversion formula may be applied, and the density estimator will be

\begin{align} \hat{f}(x) &= \frac{1}{2\pi} \int_{-\infty}^{+\infty} \hat\varphi(t)\psi_h(t) e^{-itx}dt = \frac{1}{2\pi} \int_{-\infty}^{+\infty} \frac{1}{n} \sum_{j=1}^n e^{it(x_j-x)} \psi(ht) dt \\ &= \frac{1}{nh} \sum_{j=1}^n \frac{1}{2\pi} \int_{-\infty}^{+\infty} e^{-i(ht)\frac{x-x_j}{h}} \psi(ht) d(ht) = \frac{1}{nh} \sum_{j=1}^n K\Big(\frac{x-x_j}{h}\Big), \end{align}

where K is the inverse Fourier transform of the damping function ψ. Thus the kernel density estimator coincides with the characteristic function density estimator.

Read more about this topic:  Kernel Density Estimation

Famous quotes containing the words relation to, relation and/or function:

    “Much poetry seems to be aware of its situation in time and of its relation to the metronome, the clock, and the calendar. ... The season or month is there to be felt; the day is there to be seized. Poems beginning “When” are much more numerous than those beginning “Where” of “If.” As the meter is running, the recurrent message tapped out by the passing of measured time is mortality.”
    —William Harmon (b. 1938)

    “You see, I am alive, I am alive
    I stand in good relation to the earth
    I stand in good relation to the gods
    I stand in good relation to all that is beautiful
    I stand in good relation to the daughter of Tsen-tainte
    You see, I am alive, I am alive”
    —N. Scott Momaday (b. 1934)

    “Morality and its victim, the mother—what a terrible picture! Is there indeed anything more terrible, more criminal, than our glorified sacred function of motherhood?”
    —Emma Goldman (1869–1940)