Using different line style to generate plot

I want to use different line style to gengerate the plot, since there are three values overlap in the graph. I don't know how to modify the code, can somebody help me? Thanks.

[Divyam]

Hi, please provide the code instead of an image.

[Matt J]

can somebody help me?

Yes, but first:

(1) Post your code as formatted text so we can copy-paste it.

(2) Attach N and err in a .mat file so that we can plot the data ourselves and demonstrate solutions.

[Dan]

Sorry for the inconvenience. Here is the code.

N = 10:20:500;
err = zeros(6,length(N));
for values = 1:length(N)
    err(1,values) = abs(calculate_pi("left_end",N(values))- pi);
    err(2,values) = abs(calculate_pi("right_end",N(values)) - pi);
    err(3,values) = abs(calculate_pi("midpoint",N(values)) - pi);
    err(4,values) = abs(calculate_pi("monte_carlo",N(values)) - pi);
    err(5,values) = abs(calculate_pi("leibniz",N(values)) - pi);
    err(6,values) = abs(calculate_pi("trapezoidal",N(values)) - pi);
end
loglog(N,err','-*')
legend('left-end','right-end','mid-point','monte-carlo','leibniz','trapezodial')
xlabel('N')
ylabel('Error')

[Matt J]

Sorry for the inconvenience. Here is the code.

Yes, but you still haven't attached N and err in a .mat file, as I suggested. Since we don't have calculate_pi(), we cannot regenerate them ourselves.

Answer 1

[Matt J]

N = 10:20:500;
err = rand(6,numel(N)); %fake

h = loglog(N,err');
[h.LineStyle]=deal('-','--','-.',':','--','--');
[h(end-1:end).Marker]=deal('o','x');

legend('left-end','right-end','mid-point','monte-carlo','leibniz','trapezodial',...
        location='southwest')
xlabel('N')
ylabel('Error')

Answer 2

[Chris Burschyk]

figure();

hold on

loglog(N,err(1,:)',"-")

loglog(N,err(2,:)',"--")

loglog(N,err(3,:)',"-.")

loglog(N,err(4,:)',":")

loglog(N,err(5,:)',"-^")

loglog(N,err(6,:)',"-v")

hold off

legend('left-end','right-end','mid-point','monte-carlo','leibnitz','trapezodial')

xlabel('N')

ylabel('Error')

Maybe something like this?

Answer 3

[Vinayak Agrawal]

Hi Dan,

According to me this should be the possible updates in your code-

Use different line styles for each dataset in your plot, you can modify the `loglog` function call by specifying the 'LineStyle' parameter. Here is an updated version of your code with different line styles:

N = 10:20:500;

err = zeros(6,length(N));

for values = 1:length(N)

err(1, values) = abs(calculate_pi('left_end', N(values)) - pi);

err(2, values) = abs(calculate_pi('right_end', N(values)) - pi);

err(3, values) = abs(calculate_pi('midpoint', N(values)) - pi);

err(4, values) = abs(calculate_pi('monte_carlo', N(values)) - pi);

err(5, values) = abs(calculate_pi('leibniz', N(values)) - pi);

err(6, values) = abs(calculate_pi('trapezoidal', N(values)) - pi);

end

% Specify line styles

line_styles = {'-', '--', '-.', ':', '-', '-.'};

loglog(N, err', '*', 'LineWidth', 1.5, 'Markersize', 8, 'LineStyle', line_styles);

xlabel('N');

ylabel('Error');

title('Approximation Error vs. N for Different Pi Calculation Methods');

legend('left-end', 'right-end', 'mid-point', 'monte-carlo', 'leibniz', 'trapezoidal');

Here, the `loglog` function call is modified with the addition of the `'LineStyle'` parameter, which is set to a cell array of line styles. The order of the line styles must correspond to the order of the datasets in the matrix `err`. The resulting plot will have a different line style for each dataset.

Hope it helps!