Roberto is wrigth however, you can turn your TGA data into DTG through derivation (dm/dt) and this is useful for a better determination of the mean temperature for any transformation arising within your sample. It must be quoted that you will obtained curves of profile similar to that of DTA.
I agree with Roberto Köferstein. But, it is possible that you take the derivative from TG curve. this Derived curve is similar to DSC curve. Please not that this curve is not exactly the same of DSC curve and you can not extract some DATA e.g. Cp and delta H.
I have found to get a complete thermal profile of a material you need both. If you have weight loss from TGA you have to account for this weight loss in your DSC/DTA. Similarly as others have suggested the DSC is necessary to make sense of some TGA data. It can suggest whether your seeing evaporation or decomposition.
Most answers to your questions are perfect. Take into account that there is some delay of the DSC trace with respect to the corresponding DTG trace, since the DTG immediately reflects the mass change rate, whereas the DSC signal is delayed because of the time required to transfer the heat from the sample to the thermocouples, the so-called calorimeter constant. For a precise match one must take into account this detail.