An exponential inequality for $U$-statistics of i.i.d.\ data
    
    
  
  
  
      
      
      
        
Teoriâ veroâtnostej i ee primeneniâ, Tome 66 (2021) no. 3, pp. 508-533
    
  
  
  
  
  
    
      
      
        
      
      
      
    Voir la notice de l'article provenant de la source Math-Net.Ru
            
              			We establish an exponential inequality for degenerated $U$-statistics
of order $r$ of independent and identically distributed (i.i.d.) data.
This inequality gives a control of the tail of the maxima absolute values
of the $U$-statistic by the sum of the two terms: an exponential term and one
involving the tail of $h(X_1,\dots,X_r)$. We also give a version for not
necessarily degenerated $U$-statistics having a symmetric kernel and furnish
an application to the convergence rates in the Marcinkiewicz law of large
numbers. Application to the invariance principle in Hölder spaces is also
considered.
			
            
            
            
          
        
      
                  
                    
                    
                    
                    
                    
                      
Keywords: 
$U$-statistics, exponential inequality.
                    
                  
                
                
                @article{TVP_2021_66_3_a5,
     author = {D. Giraudo},
     title = {An exponential inequality for $U$-statistics of i.i.d.\ data},
     journal = {Teori\^a vero\^atnostej i ee primeneni\^a},
     pages = {508--533},
     publisher = {mathdoc},
     volume = {66},
     number = {3},
     year = {2021},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/TVP_2021_66_3_a5/}
}
                      
                      
                    D. Giraudo. An exponential inequality for $U$-statistics of i.i.d.\ data. Teoriâ veroâtnostej i ee primeneniâ, Tome 66 (2021) no. 3, pp. 508-533. http://geodesic.mathdoc.fr/item/TVP_2021_66_3_a5/
