Firstly you have to find the lambda max of pure Tocopherols (99.9% pure) and confirm that there is no components in olive oil that has a similar lambda max value. Ones you have done with it, you can plot a standard graph of pure Tocopherol (varying concentration) by setting your spectrophotometer at Tocopherol's lambda max. Then measure your olive oil spec value at Tocopherol's lambda max value and plot it in the standard graph. Hope this helps.
You may go for FTNIR based quantification by developing the calibration model using authentic tocopherol standard dissolved in oil at wide range of concentration. Once the model calibration and validation is done, you can directly estimate the tocopherol content in oil samples.