implementing linear regression for a vector with missing values

I have two vectors x=[1 NaN 2 3 NaN 4 NaN NaN NaN 5 ]; y=[10 11 11 6 8 9 7 2 29 93 ]; I would like to know if there is a way to fit a linear regression method to the vector and estimate the missing values in x?

Answer 1

[KSSV]

x=[1 NaN 2 3 NaN 4 NaN NaN NaN 5 ]; 
y=[10 11 11 6 8 9 7 2 29 93 ];

xi = interp1(y(~isnan(x)),x(~isnan(x))) ;

x(isnan(x))= xi ;
p = polyfit(x,y,1) ;

x1 = linspace(min(x),max(x));
y1 = polyval(p,x1);

plot(x,y,'.r') ;
hold on
plot(x1,y1)

[Image Analyst]

Huh?