There is no point in trying to make something like a 3-halogenomethyl-indole. Even if it is stabile at all (I don't know) it is nevertheless probably impossible to get it with a halogenomethylation (as Rhodium pointed out).
It is much better to use gramines as the alkylating agents for the nitroacetic acid esters:
Lyttle and Weisblat. The Chemistry of Nitroacetic Acid and its esters. I. The alkylation of Alkylnitroacetates with Gramine. J. Amer. Chem. Soc. 69 (1947), 2118-2119.
Lyttle and Weisblat. The Chemistry of Nitroacetic Acid and its esters. II. The Synthesis of Ethyl-alpha-Nitro-betta-(3-indole)-propionate from Gramine and Ethyl Nitromalonate. J. Amer. Chem. Soc. 71 (1949), 3079-3081.
Gramines themselves are easily made by the aminomethylation of indoles. But unless the appropriate gramines are more easily accessible than their indole counterparts I don't see much value in this route.
Firstly, because I'm not so sure how smoothly the decarboxylation of the product goes (it probably does well but I saw no such example in the above papers as they were up to tryptophan more than anything else).
Secondly, because the 2-(3-indolyl)-nitroethanes can be prepared more easily by the electrophylic substitution on the indoles with nitroethene (NO2-CH=CH2) or 2-nitropropene (for the alpha-Me-tryptamines).
Even easier would be trying to react formaldehide, 2-nitropropane and an indole to end up with an alpha, alpha-dimethyl-tryptamine. Or one could use 2,2-dimethyl-2-nitro-ethanol instead of formaldehide+2-nitropropane. In any case it would be a cheap 2-step preparation. But who knows which ones of these tryptamines, if any, are active and interesting (though I suspect many are).