Web-Books
in the Austria-Forum
Austria-Forum
Web-Books
Naturwissenschaften
Physik
Differential Geometrical Theory of Statistics
Page - 298 -
  • User
  • Version
    • full version
    • text only version
  • Language
    • Deutsch - German
    • English

Page - 298 - in Differential Geometrical Theory of Statistics

Image of the Page - 298 -

Image of the Page - 298 - in Differential Geometrical Theory of Statistics

Text of the Page - 298 -

Entropy2016,18, 442 4.1. BasicBounds Forapairofgivenm(x)andm′(x),weonlyneedtoderiveboundsofHα(m :m′) inEquation(31) so thatLα(m :m′)≤Hα(m :m′)≤Uα(m :m′). Then theα-divergenceDα(m :m′) canbebounded byalinear transformationof therange [Lα(m :m′),Uα(m :m′)]. In thefollowingwealwaysassume without lossofgeneralityα≄1/2.OtherwisewecanboundDα(m :m′)byconsideringequivalently theboundsofD1āˆ’Ī±(m′ :m). Recall that ineachelementaryslab Ir,wehave max { kw (r)p (r)(x),wĪ“(r)pĪ“(r)(x) } ≤m(x)≤ kwĪ“(r)pĪ“(r)(x). (32) Notice thatkw (r)p (r)(x),wĪ“(r)pĪ“(r)(x), andkwĪ“(r)pĪ“(r)(x)areall singlecomponentdistributions up toascalingcoefficient. Thegeneral thinking is toboundthemulti-componentmixturem(x)by singlecomponentdistributions ineachelementary interval, so that the integral inEquation(31)canbe computedinapiecewisemanner. For theconvenienceofnotation,werewriteEquation(32)as cν(r)pν(r)(x)≤m(x)≤ cĪ“(r)pĪ“(r)(x), (33) where cν(r)pν(r)(x) := kw (r)p (r)(x) or wĪ“(r)pĪ“(r)(x), cĪ“(r)pĪ“(r)(x) := kwĪ“(r)pĪ“(r)(x). (34) (35) If1/2≤α<1, thenbothxα andx1āˆ’Ī± aremonotonically increasingonR+. Thereforewehave Aαν(r),ν′(r)(Ir)≤ ∫ Ir m(x)αm′(x)1āˆ’Ī±dx≤AαΓ(r),Γ′(r)(Ir), (36) where Aαi,j(I)= ∫ I (cipi(x)) α ( c′jp ′ j(x) )1āˆ’Ī± dx, (37) and Idenotesan interval I=(a,b)āŠ‚R. Theothercaseα>1 is similarbynoting thatxα andx1āˆ’Ī± are monotonically increasinganddecreasingonR+, respectively. Inconclusion,weobtain the following boundsofHα(m :m′): If 1/2≤α<1, Lα(m :m′)= āˆ‘ r=1 Aαν(r),ν′(r)(Ir), Uα(m :m ′)= āˆ‘ r=1 AαΓ(r),Γ′(r)(Ir); (38) ifα>1, Lα(m :m′)= āˆ‘ r=1 Aαν(r),Γ′(r)(Ir), Uα(m :m ′)= āˆ‘ r=1 AαΓ(r),ν′(r)(Ir). (39) The remaining problem is to compute the definite integral Aαi,j(I) in the above equations. Here we assume all mixture components are in the same exponential family so that pi(x)= p(x;Īøi)= h(x)exp ( Īø i t(x)āˆ’F(Īøi) ) ,whereh(x) isabasemeasure, t(x) isavectorofsufficient statistics, andthe functionF isknownas thecumulantgeneratingfunction. Then it is straightforward fromEquation(37) that Aαi,j(I)= c α i (c ′ j) 1āˆ’Ī± ∫ I h(x)exp (( αθi+(1āˆ’Ī±)θ′j ) t(x)āˆ’Ī±F(Īøi)āˆ’(1āˆ’Ī±)F(θ′j) ) dx. (40) 298
back to the  book Differential Geometrical Theory of Statistics"
Differential Geometrical Theory of Statistics
Title
Differential Geometrical Theory of Statistics
Authors
FrƩdƩric Barbaresco
Frank Nielsen
Editor
MDPI
Location
Basel
Date
2017
Language
English
License
CC BY-NC-ND 4.0
ISBN
978-3-03842-425-3
Size
17.0 x 24.4 cm
Pages
476
Keywords
Entropy, Coding Theory, Maximum entropy, Information geometry, Computational Information Geometry, Hessian Geometry, Divergence Geometry, Information topology, Cohomology, Shape Space, Statistical physics, Thermodynamics
Categories
Naturwissenschaften Physik
Web-Books
Library
Privacy
Imprint
Austria-Forum
Austria-Forum
Web-Books
Differential Geometrical Theory of Statistics