Escalating Tension and the War in Ukraine : Evidence Using Impulse Response Functions on Economic Indicators and Twitter Sentiment

In this paper, we examine the real-time effects of the escalating Ukrainian crisis on a series of economic indicators. We proxy the public’s perception of the progression of events in the armed conflict using sentiment analysis on an extended sample of 42 million tweets. The sentiment indicators are then used in a Vector Autoregressive model on 5-min data for 15 economic indicators (stock markets, commodities, interest rates, currencies and cryptocurrencies). We calculate impulse response functions and find that the effects are asymmetric across different markets and across geographic regions, noting that most indices experience a negative response within 10 to 20 minutes of “shocks” on the progression of the conflict, with crude oil registering a strong, persistent response. We also note that the US Dollar and the Chinese Renminbi appear to respond positively to the negative events of the war. Our findings provide initial evidence on the immediate effects of the war. The directions of the different impulse responses can be used as indications of expected outcomes on the examined variables over the longer term