If you mix two solutions of the same compound at equal concentration, independently of the amount of each of them, the final concentration of that compound solution does not change.
If you mix the same amount of two different compounds the concentration of them will be 50% of the original one. In your example 25 ppm each.
If you add the same quantity, in parts per million, of each of your 2 compounds, from more concentrated stock solutions, in the same diluent/mobile phase, THEN make up to the desired volume, to give 50ppm EACH, this will give you 2 peaks that should be the same height as 50ppm of each compound injected separately. However, this will only hold true if the 2 peaks are completely resolved, as any cross-over at peak base will affect the point at which the peak height can be accurately measured. A few test runs, at different concentrations (you can further dilute the 50ppm mixture in the same diluent, to preserve the 1:1 concentration ratio) will demonstrate whether the concentration:peak height relationship provides a straight line.
If you have two different compounds you will not get the same peak height because the response of detector is depending on the structure of your 2 compounds .
If you wonder if the height of the peaks of both compounds will be the same (compound 1, 50 ppm; compound 2, 50 ppm), the answer is "not necessary". It will depend on the molar extinction coefficient of each substance (how strongly a substance absorbs light at a given wavelength).
If you mixed as you state then you have diluted your standard, which was Zoe's point. Assuming that you injected the same mass using a mixed standard that you did individually, and that your chromatography resolves the two compounds, then the areas (areas, not heights) should be about the same, given experimental error. We're forced into making a lot of assumptions because you couldn't be bothered to give us any details. To quote Rod's law, "The less you tell me the more I can't help you."
It´is not about mathematix and it is, so i think, ... as Jasim Al-saadi said, it´s typically due to the detection , ) more or less, all about nm(eters) etc. Just a thing of calibration. Fake it till you make it.