No trendline display other than linear
When I plot my data (x,y) on my Mac I check the box for trendline. Only linear shows up. None of the others display, nor do the equations. Any idea why?
When I plot my data (x,y) on my Mac I check the box for trendline. Only linear shows up. None of the others display, nor do the equations. Any idea why?
First, I am assuming you are using a scatter chart because that is the only one that will create a true x-y chart and provide the true trend line equation for your x-y data.
Polynomial, moving average and linear should always make a trend line, as far as I know, but the others have restrictions. If only linear is creating a trend line, can you provide a snippet of your data that shows this problem? Please make a chart and try out the trend lines for just that snippet of data to be sure that it actually fails the same as your full data set.
Logarithmic requires all x values to be positive.
y = m*ln(x)+b
Power requires all x and y values to be positive for a line to be created
y=m*x^a
Exponential requires all x and y values to be positive for a line to be created
y=m*e^(a*x)
The equations for power and exponential are valid under other circumstances (such as if all y's are negative) but Numbers limits it.
The x = 0 in the first row is the problem. If you change it ever so slightly so it is just above zero (0.00000001 for example), you can apply the other trendlines. I'm not sure it is worth doing this, though, because none of them fit your data very well. Of the built-in trendlines, polynomial appears to be the best fit.
Logarithmic cannot have ln(0) because that ln(0) = -infinity
y = m*ln(x)+b
Power does not accept x<=0. At x=0, this curve would start at either (0,0) or (0, +/-infinity).
y=m*x^a
Exponential does not accept x<=0. It must have something to do with how it calculates the curve because, theoretically, it should work with x<=0. You can calculate the trendline equation yourself in the table and plot it. I can tell you how if you want to but it does not fit your data.
y=m*e^(a*x)
Actually, a close trendline is y = 85 + 36.6499*e^(-0.34768*x)
I don't know if that makes sense for the process you are charting but it fits your data pretty good. The 85 was trial and error and the other coefficients were calculated from there. If it was exponentially decaying to 85 degrees, that would make sense.
Here's the data
Plot time on x-axis and temperature y-axis. Then try any trendline. Other than moving average and linear non display. Nor do fir equations. Data is exponential.
Using other SW the fit is 87.03 + 34.73 * exp(-6.2909e-3*time) The SW states Rsq=1.000. So your fit is pretty good. I know from the physics the time constant is fairly long.
No trendline display other than linear