changes to sampling frequency result in wrong answer

given the following code to determine frequency, Ts=0.001,hence fs=1000

 t = 0:0.001:0.1-0.001;
 x = 10*cos(267*pi*t+1.2);
 N=max(size(x));
 t1=0;
 t2=0;
 for n=3:N
 t1=t1+x(n-1)*(x(n)+x(n-2));
 t2=t2+2*(x(n-1))^2;
 end
 r = t1/t2;
 if (r>1)
 r=1;
 end
 if (r<-1)
 r=-1;
 end
 w=acos(r)

now i want to replace x = 10*cos(267*pi*t+1.2) by the third column imported from a text file.

Fs becomes 1, the code is now changed into

 ssn = spot_num(:,3);
 ssn = ssn-mean(ssn);
 x=ssn;
 N=max(size(x));
 t1=0;
 t2=0;
 for n=3:N
 t1=t1+x(n-1)*(x(n)+x(n-2));
 t2=t2+2*(x(n-1))^2;
 end
 r = t1/t2;
 if (r>1)
 r=1;
 end
 if (r<-1)
 r=-1;
 end
 w=acos(r)

the output is not what i expected, i guess the problem lies in the unit needed to be further changed, but have no idea of doing so.

[modified covariance]

same problem lies here where fs=1000

 fs = 1000;  % samplig frequency [Hz]
 t = 0:1/fs:0.1-1/fs; % time vector
 w = 267*pi; % oscilation frequency [rad/s] (133.5 Hz)
 p = 1.2;    % phase [rad] (68.7549 deg)
 x = 10*cos(w*t+p); % wave vector
 N=length(x);   % N:number of values
 r=zeros(1,3);  % r={r0,r1,r2}
 for k=0:2,     % k=0,1,2
    n=(0:N-1-k);
    r(k+1) = sum(x(n+1).*x(n+1+k))/(N-k);
 end
 w0 = fs*acos((r(3)+sqrt(r(3)^2+8*r(2)^2))/(4*r(2))); % [rad/s]
 fprintf('w0 = %f rad/s = %f Hz\n',w0,w0/2/pi);

now i want to replace x = 10*cos(267*pi*t+1.2) by the third column imported from a text file as shown above.

Fs now Fs becomes 1.the code now is changed into

 ssn = spot_num(:,3);
 ssn = ssn-mean(ssn);
 x=ssn;
 N=max(size(x));
 N=length(x);   % N:number of values
 r=zeros(1,3);  % r={r0,r1,r2}
 for k=0:2,     % k=0,1,2
    n=(0:N-1-k);
    r(k+1) = sum(x(n+1).*x(n+1+k))/(N-k);
 end
 w0 = fs*acos((r(3)+sqrt(r(3)^2+8*r(2)^2))/(4*r(2))); % [rad/s]
 fprintf('w0 = %f rad/s = %f Hz\n',w0,w0/2/pi);

once again the output is not consistent to model answer. so what happens then?